DELCODE_Analysis_Manual
1. Load all libraries required
library(readxl)
library(tidyverse)
library('dplyr')
library('readxl')
library(ggplot2)
library(dplyr) # required by custom function outlier_removal()
library(jmv) # ancova()
library(ggplot2) # ggplot()
library(gridExtra) # grid.arrange()
2. Pre analysis: Data import and cleaning
olink_delcode<- read_excel('DELCODE_Manual only.xlsx')
olink_delcode <- data.frame(olink_delcode)
colnames(olink_delcode) [1] "Repseudonym" "Barcode_CSF" "sex" "Age" "prmdiag"
[6] "bmi" "ApoE" "E4_Positive" "AB_Ratio_Patho" "tTau_Patho"
[11] "AT" "AN" "YKL40_ng_ml" "AXL_ng_ml" "Tyro3_pg_ml"
[16] "TREM2_pg_ml" "C1q_ng_ml" "C3_ng_ml" "C4_ng_ml" "Factor_B_ng_ml"
[21] "Factor_H_ng_ml" "MIF_pg_ml" "TNFR1_ng_mL" "TNFR2_ng_mL" "ICAM1_ng_mL"
[26] "VCAM1_ng_mL" "CRP_pg_ml" DC_thesis_manual <- subset(olink_delcode, select = c(Barcode_CSF,
YKL40_ng_ml, AXL_ng_ml, Tyro3_pg_ml,
TREM2_pg_ml, C1q_ng_ml, C3_ng_ml, C4_ng_ml,
Factor_B_ng_ml, Factor_H_ng_ml, MIF_pg_ml,
TNFR1_ng_mL, TNFR2_ng_mL, ICAM1_ng_mL, VCAM1_ng_mL,
CRP_pg_ml, Age, sex, bmi, E4_Positive,
prmdiag, AB_Ratio_Patho, tTau_Patho, AT, AN))
DC_thesis_manual$A_N_cat <- factor(ifelse(DC_thesis_manual$AN == 0,
"A-T-",
ifelse(DC_thesis_manual$AN == 2,
"A+T-",
ifelse(DC_thesis_manual$AN == 1,
"A-T+",
"A+T+"))),
levels = c("A-T-", "A-T+", "A+T-", "A+T+"))
table(DC_thesis_manual$A_N_cat)
A-T- A-T+ A+T- A+T+
147 20 45 80 DC_thesis_manual$Diag <- factor(ifelse(DC_thesis_manual$prmdiag == 0 | DC_thesis_manual$prmdiag == 100,
"CN",
ifelse(DC_thesis_manual$prmdiag == 1,
"SCD",
ifelse(DC_thesis_manual$prmdiag == 2,
"MCI", "DAT"))),
levels = c('CN',"SCD", "MCI", "DAT"))
table(DC_thesis_manual$Diag)
CN SCD MCI DAT
94 94 68 37 DC_thesis_manual$clinical_N_cat <- factor(ifelse(DC_thesis_manual$Diag == "CN" &
DC_thesis_manual$tTau_Patho == 0,
"CN_T-",
ifelse(DC_thesis_manual$Diag == "CN" &
DC_thesis_manual$tTau_Patho == 1,
"CN_T+",
ifelse(DC_thesis_manual$Diag == "SCD" &
DC_thesis_manual$tTau_Patho == 0,
"SCD_T-",
ifelse(DC_thesis_manual$Diag == "SCD" &
DC_thesis_manual$tTau_Patho == 1,
"SCD_T+",
ifelse(DC_thesis_manual$Diag == "MCI" &
DC_thesis_manual$tTau_Patho == 0,
"MCI_T-",
ifelse(DC_thesis_manual$Diag == "MCI" &
DC_thesis_manual$tTau_Patho == 1,
"MCI_T+",
ifelse(DC_thesis_manual$Diag == "DAT" &
DC_thesis_manual$tTau_Patho == 0,
"DAT_T-",
ifelse(DC_thesis_manual$Diag == "DAT" &
DC_thesis_manual$tTau_Patho == 1,
"DAT_T+", 'others')))))))),
levels = c('CN_T-',"CN_T+", 'SCD_T-', 'SCD_T+', "MCI_T-", "MCI_T+","DAT_T-", "DAT_T+"))
table(DC_thesis_manual$clinical_N_cat)
CN_T- CN_T+ SCD_T- SCD_T+ MCI_T- MCI_T+ DAT_T- DAT_T+
76 18 73 21 36 31 7 30 Outlier removal function Higher than or less than 3 standard deviations
outlier_removal <- function(df) {
find.outlier <- function(df) {
if(!is.numeric(df)) {
df
}
else {
arith.mean <- mean(df, na.rm = TRUE)
st.dev <- sd(df, na.rm = TRUE)
df[which(df > (arith.mean + 3*st.dev) | df < (arith.mean - 3*st.dev))] <- NA
df
}
}
df %>% dplyr::mutate_all(find.outlier)
}Removed outliers
DC_thesis_manual_clean <- outlier_removal(DC_thesis_manual[1:16])Check for merge outlier removed columns
dim(DC_thesis_manual_clean)[1] 293 16DC_thesis_m <- merge(DC_thesis_manual_clean,
DC_thesis_manual[c(1, 17:ncol(DC_thesis_manual))],
by = "Barcode_CSF")
dim(DC_thesis_m)[1] 293 28Omit NAs to have homogenous samples
DC_thesis_m <- na.omit(DC_thesis_m)
table(DC_thesis_m$A_N_cat)
A-T- A-T+ A+T- A+T+
129 17 41 62 table(DC_thesis_m$Diag)
CN SCD MCI DAT
84 82 52 31 table(DC_thesis_m$clinical_N_cat)
CN_T- CN_T+ SCD_T- SCD_T+ MCI_T- MCI_T+ DAT_T- DAT_T+
69 15 66 16 28 24 7 24 Making BMI as character
DC_thesis_m$bmi <- as.numeric(as.character(DC_thesis_m$bmi)) Data is ready for analysis
3. AT scheme
3.1. YKL40_AT
(YKL40_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = YKL40_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "YKL40 (ng/mL)") +
scale_y_continuous(breaks = seq(0,1200, by = 200), limits = c(0,1200)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = YKL40_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = YKL40_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - YKL40_ng_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
A_N_cat 442328.374 3 147442.791 11.7518258 0.0000003
Age 370153.446 1 370153.446 29.5028247 0.0000001
sex 15863.738 1 15863.738 1.2644083 0.2619370
bmi 19605.960 1 19605.960 1.5626795 0.2124858
E4_Positive 2253.068 1 2253.068 0.1795792 0.6721131
Residuals 3023675.917 241 12546.373
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -130.68065 29.24838 241.0000 -4.4679622 0.0000729
- A+T- 16.11617 22.78354 241.0000 0.7073602 1.0000000
- A+T+ -70.93456 20.69020 241.0000 -3.4284141 0.0042834
A-T+ - A+T- 146.79681 34.22983 241.0000 4.2885640 0.0001562
- A+T+ 59.74609 32.84650 241.0000 1.8189484 0.4209620
A+T- - A+T+ -87.05073 23.10788 241.0000 -3.7671444 0.0012457
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(YKL40_ng_ml, na.rm = TRUE), Std=sd(YKL40_ng_ml, na.rm = TRUE),
Max=max(YKL40_ng_ml, na.rm = TRUE), Min=min(YKL40_ng_ml, na.rm = TRUE))NA3.2. AXL_AT
(AXL_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = AXL_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "AXL (ng/mL)") +
scale_y_continuous(breaks = seq(0,50, by = 10), limits = c(0,50)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = AXL_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = AXL_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - AXL_ng_ml
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
A_N_cat 986.559554 3 328.853185 15.60766618 < .0000001
Age 7.384877 1 7.384877 0.35049287 0.5543892
sex 1.067638 1 1.067638 0.05067104 0.8220905
bmi 74.299250 1 74.299250 3.52630885 0.0616097
E4_Positive 7.811091 1 7.811091 0.37072138 0.5431848
Residuals 5077.864722 241 21.069978
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
──────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -7.07852803 1.1986013 241.0000 -5.90565706 < .0000001
- A+T- -0.08815696 0.9336714 241.0000 -0.09441968 1.0000000
- A+T+ -3.10159674 0.8478862 241.0000 -3.65803421 0.0018729
A-T+ - A+T- 6.99037108 1.4027416 241.0000 4.98336332 0.0000072
- A+T+ 3.97693129 1.3460526 241.0000 2.95451409 0.0206510
A+T- - A+T+ -3.01343979 0.9469632 241.0000 -3.18221437 0.0099220
──────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(AXL_ng_ml, na.rm = TRUE), Std=sd(AXL_ng_ml, na.rm = TRUE),
Max=max(AXL_ng_ml, na.rm = TRUE), Min=min(AXL_ng_ml, na.rm = TRUE))3.3. Tyro3_AT
(Tyro3_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = Tyro3_pg_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Tyro3 (pg/mL)") +
scale_y_continuous(breaks = seq(0,14000, by = 2000), limits = c(0,14000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Tyro3_pg_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Tyro3_pg_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Tyro3_pg_ml
───────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────
A_N_cat 6.898650e+7 3 2.299550e+7 16.9731533 < .0000001
Age 379212.6 1 379212.6 0.2798997 0.5972539
sex 459956.7 1 459956.7 0.3394975 0.5606641
bmi 1776868.4 1 1776868.4 1.3115201 0.2532556
E4_Positive 746948.7 1 746948.7 0.5513286 0.4584984
Residuals 3.265107e+8 241 1354816.0
───────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -2007.5910 303.9365 241.0000 -6.6052978 < .0000001
- A+T- 123.5141 236.7566 241.0000 0.5216920 1.0000000
- A+T+ -526.4993 215.0036 241.0000 -2.4487933 0.0902824
A-T+ - A+T- 2131.1051 355.7016 241.0000 5.9912725 < .0000001
- A+T+ 1481.0917 341.3266 241.0000 4.3392215 0.0001263
A+T- - A+T+ -650.0134 240.1271 241.0000 -2.7069553 0.0436571
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(Tyro3_pg_ml, na.rm = TRUE), Std=sd(Tyro3_pg_ml, na.rm = TRUE),
Max=max(Tyro3_pg_ml, na.rm = TRUE), Min=min(Tyro3_pg_ml, na.rm = TRUE))NA3.4. TREM2_AT
(TREM2_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = TREM2_pg_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TREM2 (pg/mL)") +
scale_y_continuous(breaks = seq(0,14000, by = 2000), limits = c(0,14000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TREM2_pg_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TREM2_pg_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TREM2_pg_ml
─────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────────
A_N_cat 5.101179e+7 3 1.700393e+7 12.891782625 < .0000001
Age 7200216.833 1 7200216.833 5.458951863 0.0202887
sex 4184.927 1 4184.927 0.003172865 0.9551270
bmi 833633.810 1 833633.810 0.632031916 0.4273932
E4_Positive 3.440095e+7 1 3.440095e+7 26.081591277 0.0000007
Residuals 3.178728e+8 241 1318974.231
─────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -1691.1851 299.8892 241.0000 -5.6393661 0.0000003
- A+T- -165.6639 233.6039 241.0000 -0.7091655 1.0000000
- A+T+ -684.6612 212.1405 241.0000 -3.2273941 0.0085359
A-T+ - A+T- 1525.5212 350.9650 241.0000 4.3466479 0.0001224
- A+T+ 1006.5239 336.7814 241.0000 2.9886563 0.0185517
A+T- - A+T+ -518.9973 236.9295 241.0000 -2.1905133 0.1766732
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(TREM2_pg_ml, na.rm = TRUE), Std=sd(TREM2_pg_ml, na.rm = TRUE),
Max=max(TREM2_pg_ml, na.rm = TRUE), Min=min(TREM2_pg_ml, na.rm = TRUE))3.5. C1q_AT
(C1q_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = C1q_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C1q (ng/mL)") +
scale_y_continuous(breaks = seq(0,600, by = 100), limits = c(0,600)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C1q_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C1q_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C1q_ng_ml
───────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────
A_N_cat 74226.91726 3 24742.30575 7.79243479 0.0000550
Age 24016.88597 1 24016.88597 7.56396836 0.0064063
sex 51991.43830 1 51991.43830 16.37437904 0.0000701
bmi 51.40037 1 51.40037 0.01618823 0.8988621
E4_Positive 3572.97375 1 3572.97375 1.12528579 0.2898450
Residuals 765215.98757 241 3175.17007
───────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -69.018590 14.71385 241.0000 -4.6907242 0.0000274
- A+T- -2.429943 11.46161 241.0000 -0.2120072 1.0000000
- A+T+ -15.868429 10.40852 241.0000 -1.5245614 0.7720783
A-T+ - A+T- 66.588647 17.21984 241.0000 3.8669722 0.0008508
- A+T+ 53.150161 16.52393 241.0000 3.2165559 0.0088510
A+T- - A+T+ -13.438486 11.62477 241.0000 -1.1560212 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(C1q_ng_ml, na.rm = TRUE), Std=sd(C1q_ng_ml, na.rm = TRUE),
Max=max(C1q_ng_ml, na.rm = TRUE), Min=min(C1q_ng_ml, na.rm = TRUE))3.6. C3_AT
(C3_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = C3_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C3 (ng/mL)") +
scale_y_continuous(breaks = seq(0,5000, by = 500), limits = c(0, 5000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C3_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C3_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C3_ng_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
A_N_cat 397849.49 3 132616.50 0.27540989 0.8431128
Age 698109.99 1 698109.99 1.44979243 0.2297425
sex 900828.01 1 900828.01 1.87078493 0.1726593
bmi 1387099.44 1 1387099.44 2.88064391 0.0909411
E4_Positive 47539.05 1 47539.05 0.09872622 0.7536354
Residuals 1.160473e+8 241 481524.09
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
─────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
─────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -81.04673 181.1973 241.0000 -0.44728442 1.0000000
- A+T- -119.05666 141.1468 241.0000 -0.84349528 1.0000000
- A+T+ -63.47687 128.1783 241.0000 -0.49522319 1.0000000
A-T+ - A+T- -38.00993 212.0580 241.0000 -0.17924310 1.0000000
- A+T+ 17.56985 203.4881 241.0000 0.08634340 1.0000000
A+T- - A+T+ 55.57979 143.1562 241.0000 0.38824584 1.0000000
─────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(C3_ng_ml, na.rm = TRUE), Std=sd(C3_ng_ml, na.rm = TRUE),
Max=max(C3_ng_ml, na.rm = TRUE), Min=min(C3_ng_ml, na.rm = TRUE))3.7. C4_AT
(C4_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = C4_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C4 (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C4_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C4_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C4_ng_ml
────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────
A_N_cat 335243.31 3 111747.77 3.140854 0.0260016
Age 132323.55 1 132323.55 3.719170 0.0549657
sex 80161.83 1 80161.83 2.253080 0.1346578
bmi 110008.14 1 110008.14 3.091959 0.0799493
E4_Positive 107046.69 1 107046.69 3.008723 0.0840961
Residuals 8574487.02 241 35578.78
────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -128.13673 49.25363 241.0000 -2.6015693 0.0591252
- A+T- 28.29967 38.36698 241.0000 0.7376050 1.0000000
- A+T+ -34.64201 34.84184 241.0000 -0.9942648 1.0000000
A-T+ - A+T- 156.43641 57.64229 241.0000 2.7139175 0.0427772
- A+T+ 93.49472 55.31279 241.0000 1.6902913 0.5535912
A+T- - A+T+ -62.94169 38.91317 241.0000 -1.6174907 0.6424828
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(C4_ng_ml, na.rm = TRUE), Std=sd(C4_ng_ml, na.rm = TRUE),
Max=max(C4_ng_ml, na.rm = TRUE), Min=min(C4_ng_ml, na.rm = TRUE))3.8. FactorB_AT
(FB_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = Factor_B_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Factor B (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Factor_B_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Factor_B_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Factor_B_ng_ml
─────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────
A_N_cat 288200.40 3 96066.80 2.1965089 0.0890931
Age 21339.27 1 21339.27 0.4879094 0.4855348
sex 48461.88 1 48461.88 1.1080514 0.2935602
bmi 351605.64 1 351605.64 8.0392490 0.0049661
E4_Positive 112323.12 1 112323.12 2.5681999 0.1103416
Residuals 1.054041e+7 241 43736.13
─────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -105.22191 54.60883 241.0000 -1.9268296 0.3310600
- A+T- 58.71830 42.53850 241.0000 1.3803566 1.0000000
- A+T+ 11.39858 38.63009 241.0000 0.2950701 1.0000000
A-T+ - A+T- 163.94021 63.90956 241.0000 2.5651908 0.0655104
- A+T+ 116.62049 61.32678 241.0000 1.9016243 0.3504774
A+T- - A+T+ -47.31972 43.14408 241.0000 -1.0967836 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(Factor_B_ng_ml, na.rm = TRUE), Std=sd(Factor_B_ng_ml, na.rm = TRUE),
Max=max(Factor_B_ng_ml, na.rm = TRUE), Min=min(Factor_B_ng_ml, na.rm = TRUE))3.9. FactorH_AT
(FH_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = Factor_H_ng_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Factor H (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Factor_H_ng_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Factor_H_ng_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Factor_H_ng_ml
────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────
A_N_cat 335622.53 3 111874.18 4.225489 0.0061891
Age 62897.00 1 62897.00 2.375621 0.1245546
sex 137237.08 1 137237.08 5.183447 0.0236805
bmi 91109.31 1 91109.31 3.441200 0.0648107
E4_Positive 67810.31 1 67810.31 2.561196 0.1108257
Residuals 6380722.79 241 26476.03
────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -124.35378 42.48828 241.0000 -2.9267788 0.0225143
- A+T- 43.02745 33.09698 241.0000 1.3000415 1.0000000
- A+T+ -19.43513 30.05605 241.0000 -0.6466295 1.0000000
A-T+ - A+T- 167.38124 49.72469 241.0000 3.3661597 0.0053216
- A+T+ 104.91866 47.71516 241.0000 2.1988536 0.1730347
A+T- - A+T+ -62.46258 33.56815 241.0000 -1.8607691 0.3839680
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(Factor_H_ng_ml, na.rm = TRUE), Std=sd(Factor_H_ng_ml, na.rm = TRUE),
Max=max(Factor_H_ng_ml, na.rm = TRUE), Min=min(Factor_H_ng_ml, na.rm = TRUE))3.10. MIF_AT
(MIF_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = MIF_pg_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "MIF (pg/mL)") +
scale_y_continuous(breaks = seq(0,30000, by = 5000), limits = c(0, 30000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = MIF_pg_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = MIF_pg_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - MIF_pg_ml
─────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────────
A_N_cat 5.926305e+8 3 1.975435e+8 19.735678143 < .0000001
Age 4.176001e+7 1 4.176001e+7 4.172053426 0.0421843
sex 19385.83 1 19385.83 0.001936751 0.9649340
bmi 931325.55 1 931325.55 0.093044525 0.7606050
E4_Positive 8495250.42 1 8495250.42 0.848722059 0.3578354
Residuals 2.412280e+9 241 1.000946e+7
─────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -4987.3513 826.1293 241.0000 -6.037010 < .0000001
- A+T- 826.3967 643.5279 241.0000 1.284166 1.0000000
- A+T+ -2253.7925 584.4009 241.0000 -3.856586 0.0008856
A-T+ - A+T- 5813.7480 966.8319 241.0000 6.013194 < .0000001
- A+T+ 2733.5589 927.7593 241.0000 2.946409 0.0211803
A+T- - A+T+ -3080.1891 652.6891 241.0000 -4.719228 0.0000241
───────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(MIF_pg_ml, na.rm = TRUE), Std=sd(MIF_pg_ml, na.rm = TRUE),
Max=max(MIF_pg_ml, na.rm = TRUE), Min=min(MIF_pg_ml, na.rm = TRUE))3.11. TNFR1_AT
(TNFR1_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = TNFR1_ng_mL,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TNFR1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,3, by = 0.5), limits = c(0, 3)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TNFR1_ng_mL ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TNFR1_ng_mL ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TNFR1_ng_mL
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
A_N_cat 1.69555985 3 0.56518662 12.72789827 < .0000001
Age 0.10156776 1 0.10156776 2.28728718 0.1317479
sex 6.712165e-4 1 6.712165e-4 0.01511567 0.9022525
bmi 0.03300424 1 0.03300424 0.74324941 0.3894790
E4_Positive 0.01018089 1 0.01018089 0.22927181 0.6324984
Residuals 10.70168631 241 0.04440534
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
──────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -0.24412049 0.05502503 241.0000 -4.4365355 0.0000835
- A+T- 0.01573525 0.04286271 241.0000 0.3671081 1.0000000
- A+T+ -0.15698647 0.03892451 241.0000 -4.0331012 0.0004435
A-T+ - A+T- 0.25985574 0.06439664 241.0000 4.0352374 0.0004397
- A+T+ 0.08713402 0.06179418 241.0000 1.4100684 0.9588525
A+T- - A+T+ -0.17272172 0.04347290 241.0000 -3.9730892 0.0005625
──────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(TNFR1_ng_mL, na.rm = TRUE), Std=sd(TNFR1_ng_mL, na.rm = TRUE),
Max=max(TNFR1_ng_mL, na.rm = TRUE), Min=min(TNFR1_ng_mL, na.rm = TRUE))3.12. TNFR2_AT
(TNFR2_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = TNFR2_ng_mL,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TNFR2 (ng/mL)") +
scale_y_continuous(breaks = seq(0,5, by = 1), limits = c(0, 5)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TNFR2_ng_mL ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TNFR2_ng_mL ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TNFR2_ng_mL
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
A_N_cat 4.94904525 3 1.64968175 10.3923057 0.0000019
Age 1.44539971 1 1.44539971 9.1054142 0.0028220
sex 0.07826978 1 0.07826978 0.4930669 0.4832400
bmi 0.04539024 1 0.04539024 0.2859396 0.5933274
E4_Positive 0.34420163 1 0.34420163 2.1683264 0.1421842
Residuals 38.25650565 241 0.15874069
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
──────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -0.41268541 0.10403676 241.0000 -3.9667269 0.0005768
- A+T- -0.02632884 0.08104125 241.0000 -0.3248819 1.0000000
- A+T+ -0.29824323 0.07359523 241.0000 -4.0524806 0.0004104
A-T+ - A+T- 0.38635658 0.12175583 241.0000 3.1732080 0.0102221
- A+T+ 0.11444218 0.11683531 241.0000 0.9795171 1.0000000
A+T- - A+T+ -0.27191439 0.08219496 241.0000 -3.3081640 0.0064958
──────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(TNFR2_ng_mL, na.rm = TRUE), Std=sd(TNFR2_ng_mL, na.rm = TRUE),
Max=max(TNFR2_ng_mL, na.rm = TRUE), Min=min(TNFR2_ng_mL, na.rm = TRUE))3.13. ICAM1_AT
(ICAM1_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = ICAM1_ng_mL,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "ICAM1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,20, by = 5), limits = c(0, 20)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = ICAM1_ng_mL ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = ICAM1_ng_mL ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - ICAM1_ng_mL
─────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────
A_N_cat 40.273775 3 13.424592 5.5720826 0.0010341
Age 1.212850 1 1.212850 0.5034120 0.4786906
sex 4.788841 1 4.788841 1.9876821 0.1598727
bmi 8.321434 1 8.321434 3.4539390 0.0643204
E4_Positive 2.594765 1 2.594765 1.0769971 0.3004104
Residuals 580.631479 241 2.409259
─────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
─────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
─────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -1.4069534 0.4053073 241.0000 -3.4713248 0.0036816
- A+T- 0.1845634 0.3157212 241.0000 0.5845772 1.0000000
- A+T+ -0.5054220 0.2867129 241.0000 -1.7628154 0.4751944
A-T+ - A+T- 1.5915168 0.4743374 241.0000 3.3552418 0.0055263
- A+T+ 0.9015314 0.4551680 241.0000 1.9806562 0.2926016
A+T- - A+T+ -0.6899854 0.3202158 241.0000 -2.1547511 0.1930353
─────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(ICAM1_ng_mL, na.rm = TRUE), Std=sd(ICAM1_ng_mL, na.rm = TRUE),
Max=max(ICAM1_ng_mL, na.rm = TRUE), Min=min(ICAM1_ng_mL, na.rm = TRUE))3.14. VCAM1_AT
(VCAM1_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = VCAM1_ng_mL,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "VCAM1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,40, by = 10), limits = c(0, 40)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = VCAM1_ng_mL ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = VCAM1_ng_mL ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - VCAM1_ng_mL
─────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────
A_N_cat 110.229143 3 36.743048 4.1914719 0.0064751
Age 7.907160 1 7.907160 0.9020112 0.3431944
sex 31.331783 1 31.331783 3.5741806 0.0598841
bmi 3.182517 1 3.182517 0.3630464 0.5473857
E4_Positive 14.677502 1 14.677502 1.6743395 0.1969169
Residuals 2112.640779 241 8.766144
─────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
─────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
─────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -2.3539300 0.7731202 241.0000 -3.0447143 0.0155252
- A+T- 0.2898366 0.6022355 241.0000 0.4812678 1.0000000
- A+T+ -0.8183357 0.5469024 241.0000 -1.4963104 0.8152872
A-T+ - A+T- 2.6437666 0.9047945 241.0000 2.9219526 0.0228538
- A+T+ 1.5355943 0.8682290 241.0000 1.7686512 0.4693023
A+T- - A+T+ -1.1081723 0.6108089 241.0000 -1.8142701 0.4252776
─────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(VCAM1_ng_mL, na.rm = TRUE), Std=sd(VCAM1_ng_mL, na.rm = TRUE),
Max=max(VCAM1_ng_mL, na.rm = TRUE), Min=min(VCAM1_ng_mL, na.rm = TRUE))3.15. CRP_AT
(CRP_A_T_plot <- ggplot(DC_thesis_m,
aes(x = A_N_cat, y = CRP_pg_ml,
colour = A_N_cat, fill = A_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "CRP (pg/mL)") +
scale_y_continuous(breaks = seq(0,50000, by = 12500), limits = c(0, 50000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_brewer(palette = "Accent") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = CRP_pg_ml ~ A_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = CRP_pg_ml ~ A_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - CRP_pg_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
A_N_cat 2.539068e+7 3 8463560.9 0.24941583 0.8617091
Age 876612.3 1 876612.3 0.02583321 0.8724430
sex 6029127.6 1 6029127.6 0.17767461 0.6737550
bmi 2.402532e+8 1 2.402532e+8 7.08011105 0.0083181
E4_Positive 1.979741e+7 1 1.979741e+7 0.58341725 0.4457231
Residuals 8.177982e+9 241 3.393354e+7
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - A_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────
A_N_cat A_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────
A-T- - A-T+ -1068.6582 1521.098 241.0000 -0.7025570 1.0000000
- A+T- -659.7890 1184.886 241.0000 -0.5568375 1.0000000
- A+T+ -122.9834 1076.019 241.0000 -0.1142948 1.0000000
A-T+ - A+T- 408.8692 1780.165 241.0000 0.2296806 1.0000000
- A+T+ 945.6749 1708.223 241.0000 0.5536016 1.0000000
A+T- - A+T+ 536.8056 1201.754 241.0000 0.4466851 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(A_N_cat)%>%
summarise(Median=median(CRP_pg_ml, na.rm = TRUE), Std=sd(CRP_pg_ml, na.rm = TRUE),
Max=max(CRP_pg_ml, na.rm = TRUE), Min=min(CRP_pg_ml, na.rm = TRUE))4. Diagnosis scheme
4.1. YKL40_Diag
(YKL40_diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = YKL40_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "YKL40 (ng/mL)") +
scale_y_continuous(breaks = seq(0,1200, by = 200), limits = c(0,1200)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = YKL40_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = YKL40_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - YKL40_ng_ml
───────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────
Diag 70698.1463 3 23566.0488 1.67272626 0.1734952
Age 393192.5943 1 393192.5943 27.90894581 0.0000003
sex 7118.9274 1 7118.9274 0.50530392 0.4778662
bmi 77948.7345 1 77948.7345 5.53282803 0.0194683
E4_Positive 287.8757 1 287.8757 0.02043352 0.8864524
Residuals 3395306.1454 241 14088.4072
───────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
───────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 14.371759 19.12834 241.0000 0.75133311 1.0000000
- MCI -33.157254 22.39292 241.0000 -1.48070222 0.8399525
- DAT -1.517767 27.51219 241.0000 -0.05516706 1.0000000
SCD - MCI -47.529012 21.26475 241.0000 -2.23510811 0.1579628
- DAT -15.889525 26.22770 241.0000 -0.60583000 1.0000000
MCI - DAT 31.639487 28.10668 241.0000 1.12569293 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(YKL40_ng_ml, na.rm = TRUE), Std=sd(YKL40_ng_ml, na.rm = TRUE),
Max=max(YKL40_ng_ml, na.rm = TRUE), Min=min(YKL40_ng_ml, na.rm = TRUE))4.2. AXL_Diag
(AXL_diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = AXL_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "AXL (ng/mL)") +
scale_y_continuous(breaks = seq(0,50, by = 10), limits = c(0,50)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = AXL_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = AXL_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - AXL_ng_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 132.135381 3 44.045127 1.78933895 0.1498046
Age 15.893487 1 15.893487 0.64567496 0.4224551
sex 3.688896 1 3.688896 0.14986187 0.6990090
bmi 179.429770 1 179.429770 7.28935752 0.0074280
E4_Positive 2.224020 1 2.224020 0.09035108 0.7639904
Residuals 5932.288896 241 24.615307
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
────────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 1.62198414 0.7995563 241.0000 2.02860525 0.2615894
- MCI 1.79898890 0.9360143 241.0000 1.92196725 0.3347339
- DAT 1.54153267 1.1499974 241.0000 1.34046620 1.0000000
SCD - MCI 0.17700476 0.8888571 241.0000 0.19913749 1.0000000
- DAT -0.08045147 1.0963061 241.0000 -0.07338413 1.0000000
MCI - DAT -0.25745623 1.1748466 241.0000 -0.21914029 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(AXL_ng_ml, na.rm = TRUE), Std=sd(AXL_ng_ml, na.rm = TRUE),
Max=max(AXL_ng_ml, na.rm = TRUE), Min=min(AXL_ng_ml, na.rm = TRUE))4.3. Tyro3_Diag
(Tyro3_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = Tyro3_pg_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Tyro3 (pg/mL)") +
scale_y_continuous(breaks = seq(0,10000, by = 1000), limits = c(0,10000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Tyro3_pg_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Tyro3_pg_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Tyro3_pg_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 1.308245e+7 3 4360816.68 2.74821240 0.0435273
Age 5002869.50 1 5002869.50 3.15283788 0.0770569
sex 264779.60 1 264779.60 0.16686567 0.6832753
bmi 7157411.91 1 7157411.91 4.51064321 0.0347035
E4_Positive 92043.06 1 92043.06 0.05800608 0.8098804
Residuals 3.824147e+8 241 1586782.99
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 394.70910 203.0043 241.0000 1.9443388 0.3181093
- MCI 363.63639 237.6505 241.0000 1.5301313 0.7637745
- DAT 793.78504 291.9799 241.0000 2.7186287 0.0421909
SCD - MCI -31.07271 225.6774 241.0000 -0.1376864 1.0000000
- DAT 399.07594 278.3479 241.0000 1.4337307 0.9176715
MCI - DAT 430.14865 298.2891 241.0000 1.4420531 0.9035139
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(Tyro3_pg_ml, na.rm = TRUE), Std=sd(Tyro3_pg_ml, na.rm = TRUE),
Max=max(Tyro3_pg_ml, na.rm = TRUE), Min=min(Tyro3_pg_ml, na.rm = TRUE))4.4. TREM2_Diag
(TREM2_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = TREM2_pg_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TREM2 (pg/mL)") +
scale_y_continuous(breaks = seq(0,14000, by = 2000), limits = c(0,14000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TREM2_pg_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TREM2_pg_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TREM2_pg_ml
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
Diag 1.402052e+7 3 4673505.81 3.17393343 0.0248934
Age 1.940661e+7 1 1.940661e+7 13.17967679 0.0003456
sex 77730.65 1 77730.65 0.05278947 0.8184735
bmi 3324965.16 1 3324965.16 2.25809456 0.1342267
E4_Positive 5.686691e+7 1 5.686691e+7 38.62020959 < .0000001
Residuals 3.548641e+8 241 1472464.98
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 594.6119 195.5550 241.0000 3.0406377 0.0157290
- MCI 403.4644 228.9298 241.0000 1.7623934 0.4756228
- DAT 270.4583 281.2657 241.0000 0.9615759 1.0000000
SCD - MCI -191.1475 217.3961 241.0000 -0.8792591 1.0000000
- DAT -324.1536 268.1339 241.0000 -1.2089245 1.0000000
MCI - DAT -133.0061 287.3433 241.0000 -0.4628822 1.0000000
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(TREM2_pg_ml, na.rm = TRUE), Std=sd(TREM2_pg_ml, na.rm = TRUE),
Max=max(TREM2_pg_ml, na.rm = TRUE), Min=min(TREM2_pg_ml, na.rm = TRUE))4.5. C1q_Diag
(C1q_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = C1q_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C1q (ng/mL)") +
scale_y_continuous(breaks = seq(0,600, by = 100), limits = c(0,600)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C1q_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C1q_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C1q_ng_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 1050.6810 3 350.2270 0.1006745 0.9595653
Age 29206.5871 1 29206.5871 8.3955782 0.0041077
sex 50709.8326 1 50709.8326 14.5767927 0.0001712
bmi 876.5369 1 876.5369 0.2519649 0.6161532
E4_Positive 3290.8566 1 3290.8566 0.9459730 0.3317216
Residuals 838392.2238 241 3478.8059
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
────────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 2.7287344 9.505206 241.0000 0.28707789 1.0000000
- MCI -0.5093432 11.127433 241.0000 -0.04577365 1.0000000
- DAT -4.0449785 13.671285 241.0000 -0.29587405 1.0000000
SCD - MCI -3.2380776 10.566822 241.0000 -0.30643816 1.0000000
- DAT -6.7737130 13.032997 241.0000 -0.51973563 1.0000000
MCI - DAT -3.5356353 13.966695 241.0000 -0.25314760 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(C1q_ng_ml, na.rm = TRUE), Std=sd(C1q_ng_ml, na.rm = TRUE),
Max=max(C1q_ng_ml, na.rm = TRUE), Min=min(C1q_ng_ml, na.rm = TRUE))4.6. C3_Diag
(C3_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = C3_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C3 (ng/mL)") +
scale_y_continuous(breaks = seq(0,5000, by = 500), limits = c(0, 5000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C3_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C3_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C3_ng_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 1127900.40 3 375966.80 0.78572802 0.5028994
Age 499409.73 1 499409.73 1.04370977 0.3079847
sex 779974.29 1 779974.29 1.63005790 0.2029239
bmi 1406716.13 1 1406716.13 2.93987736 0.0877028
E4_Positive 43123.63 1 43123.63 0.09012349 0.7642790
Residuals 1.153173e+8 241 478494.84
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD -44.22173 111.4769 241.0000 -0.3966895 1.0000000
- MCI -116.76773 130.5024 241.0000 -0.8947554 1.0000000
- DAT -230.12030 160.3367 241.0000 -1.4352319 0.9151053
SCD - MCI -72.54600 123.9276 241.0000 -0.5853903 1.0000000
- DAT -185.89857 152.8508 241.0000 -1.2162090 1.0000000
MCI - DAT -113.35257 163.8012 241.0000 -0.6920129 1.0000000
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(C3_ng_ml, na.rm = TRUE), Std=sd(C3_ng_ml, na.rm = TRUE),
Max=max(C3_ng_ml, na.rm = TRUE), Min=min(C3_ng_ml, na.rm = TRUE))NA4.7. C4_Diag
(C4_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = C4_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C4 (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C4_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C4_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C4_ng_ml
────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────
Diag 137443.33 3 45814.44 1.258655 0.2892303
Age 54098.14 1 54098.14 1.486232 0.2239939
sex 67726.80 1 67726.80 1.860650 0.1738223
bmi 59909.16 1 59909.16 1.645877 0.2007534
E4_Positive 66351.20 1 66351.20 1.822859 0.1782400
Residuals 8772287.00 241 36399.53
────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD -14.49953 30.74640 241.0000 -0.4715846 1.0000000
- MCI -43.79650 35.99381 241.0000 -1.2167787 1.0000000
- DAT -78.09130 44.22238 241.0000 -1.7658772 0.4720956
SCD - MCI -29.29696 34.18040 241.0000 -0.8571275 1.0000000
- DAT -63.59176 42.15772 241.0000 -1.5084252 0.7965333
MCI - DAT -34.29480 45.17794 241.0000 -0.7591050 1.0000000
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(C4_ng_ml, na.rm = TRUE), Std=sd(C4_ng_ml, na.rm = TRUE),
Max=max(C4_ng_ml, na.rm = TRUE), Min=min(C4_ng_ml, na.rm = TRUE))4.8. FactorB_Diag
(FB_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = Factor_B_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Factor B (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Factor_B_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Factor_B_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Factor_B_ng_ml
─────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────
Diag 157040.12 3 52346.71 1.1821653 0.3171568
Age 147447.40 1 147447.40 3.3298598 0.0692706
sex 25204.62 1 25204.62 0.5692054 0.4513114
bmi 273820.80 1 273820.80 6.1837974 0.0135706
E4_Positive 21477.68 1 21477.68 0.4850384 0.4868201
Residuals 1.067157e+7 241 44280.36
─────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD -39.355753 33.91189 241.0000 -1.1605296 1.0000000
- MCI -64.423223 39.69954 241.0000 -1.6227701 0.6356755
- DAT -72.522748 48.77529 241.0000 -1.4868749 0.8301298
SCD - MCI -25.067470 37.69944 241.0000 -0.6649295 1.0000000
- DAT -33.166995 46.49805 241.0000 -0.7132985 1.0000000
MCI - DAT -8.099525 49.82922 241.0000 -0.1625457 1.0000000
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(Factor_B_ng_ml, na.rm = TRUE), Std=sd(Factor_B_ng_ml, na.rm = TRUE),
Max=max(Factor_B_ng_ml, na.rm = TRUE), Min=min(Factor_B_ng_ml, na.rm = TRUE))4.9. FactorH_Diag
(FH_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = Factor_H_ng_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Factor H (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Factor_H_ng_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Factor_H_ng_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Factor_H_ng_ml
─────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────
Diag 73524.93 3 24508.31 0.8891559 0.4473599
Age 12422.24 1 12422.24 0.4506759 0.5026571
sex 108181.14 1 108181.14 3.9247871 0.0487160
bmi 45003.39 1 45003.39 1.6327126 0.2025577
E4_Positive 25799.03 1 25799.03 0.9359828 0.3342830
Residuals 6642820.39 241 27563.57
─────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD -19.78613 26.75557 241.0000 -0.7395145 1.0000000
- MCI -40.74155 31.32186 241.0000 -1.3007385 1.0000000
- DAT -53.73278 38.48238 241.0000 -1.3962956 0.9834614
SCD - MCI -20.95542 29.74384 241.0000 -0.7045298 1.0000000
- DAT -33.94665 36.68571 241.0000 -0.9253372 1.0000000
MCI - DAT -12.99123 39.31391 241.0000 -0.3304486 1.0000000
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(Factor_H_ng_ml, na.rm = TRUE), Std=sd(Factor_H_ng_ml, na.rm = TRUE),
Max=max(Factor_H_ng_ml, na.rm = TRUE), Min=min(Factor_H_ng_ml, na.rm = TRUE))4.10. MIF_Diag
(MIF_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = MIF_pg_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "MIF (pg/mL)") +
scale_y_continuous(breaks = seq(0,30000, by = 5000), limits = c(0, 30000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = MIF_pg_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = MIF_pg_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - MIF_pg_ml
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 1.681487e+7 3 5604956 0.45205857 0.7160750
Age 5.235469e+7 1 5.235469e+7 4.22258192 0.0409676
sex 1061769 1 1061769 0.08563529 0.7700529
bmi 3.313985e+7 1 3.313985e+7 2.67284039 0.1033792
E4_Positive 2.561969e+7 1 2.561969e+7 2.06631473 0.1518821
Residuals 2.988096e+9 241 1.239874e+7
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
──────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 132.5595 567.4599 241.0000 0.2336015 1.0000000
- MCI -524.2356 664.3067 241.0000 -0.7891470 1.0000000
- DAT 258.4743 816.1744 241.0000 0.3166900 1.0000000
SCD - MCI -656.7951 630.8383 241.0000 -1.0411465 1.0000000
- DAT 125.9148 778.0687 241.0000 0.1618300 1.0000000
MCI - DAT 782.7099 833.8103 241.0000 0.9387145 1.0000000
──────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(MIF_pg_ml, na.rm = TRUE), Std=sd(MIF_pg_ml, na.rm = TRUE),
Max=max(MIF_pg_ml, na.rm = TRUE), Min=min(MIF_pg_ml, na.rm = TRUE))4.11. TNFR1_Diag
(TNFR1_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = TNFR1_ng_mL,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TNFR1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,3, by = 0.5), limits = c(0, 3)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TNFR1_ng_mL ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TNFR1_ng_mL ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TNFR1_ng_mL
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 0.092727695 3 0.030909232 0.60539751 0.6120822
Age 0.152850380 1 0.152850380 2.99377352 0.0848653
sex 0.004718956 1 0.004718956 0.09242689 0.7613766
bmi 0.005622870 1 0.005622870 0.11013121 0.7402832
E4_Positive 0.095175902 1 0.095175902 1.86414385 0.1734204
Residuals 12.304518468 241 0.051056093
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
────────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 0.01318289 0.03641415 241.0000 0.3620266 1.0000000
- MCI -0.04022735 0.04262885 241.0000 -0.9436648 1.0000000
- DAT -0.01616199 0.05237428 241.0000 -0.3085864 1.0000000
SCD - MCI -0.05341024 0.04048117 241.0000 -1.3193847 1.0000000
- DAT -0.02934488 0.04992901 241.0000 -0.5877320 1.0000000
MCI - DAT 0.02406536 0.05350598 241.0000 0.4497696 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(TNFR1_ng_mL, na.rm = TRUE), Std=sd(TNFR1_ng_mL, na.rm = TRUE),
Max=max(TNFR1_ng_mL, na.rm = TRUE), Min=min(TNFR1_ng_mL, na.rm = TRUE))4.12. TNFR2_Diag
(TNFR2_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = TNFR2_ng_mL,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TNFR2 (ng/mL)") +
scale_y_continuous(breaks = seq(0,5, by = 1), limits = c(0, 5)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TNFR2_ng_mL ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TNFR2_ng_mL ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TNFR2_ng_mL
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 0.30447437 3 0.10149146 0.5701358 0.6351801
Age 1.82995478 1 1.82995478 10.2799076 0.0015269
sex 0.05130440 1 0.05130440 0.2882063 0.5918675
bmi 0.05022237 1 0.05022237 0.2821279 0.5957991
E4_Positive 1.14753435 1 1.14753435 6.4463599 0.0117481
Residuals 42.90107653 241 0.17801277
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
────────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD -0.009591245 0.06799422 241.0000 -0.1410597 1.0000000
- MCI -0.088323307 0.07959860 241.0000 -1.1096087 1.0000000
- DAT -0.075014599 0.09779571 241.0000 -0.7670541 1.0000000
SCD - MCI -0.078732062 0.07558835 241.0000 -1.0415899 1.0000000
- DAT -0.065423354 0.09322980 241.0000 -0.7017429 1.0000000
MCI - DAT 0.013308708 0.09990889 241.0000 0.1332085 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(TNFR2_ng_mL, na.rm = TRUE), Std=sd(TNFR2_ng_mL, na.rm = TRUE),
Max=max(TNFR2_ng_mL, na.rm = TRUE), Min=min(TNFR2_ng_mL, na.rm = TRUE))4.13. ICAM1_Diag
(ICAM1_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = ICAM1_ng_mL,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "ICAM1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,20, by = 5), limits = c(0, 20)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = ICAM1_ng_mL ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = ICAM1_ng_mL ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - ICAM1_ng_mL
──────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────
Diag 9.3118490 3 3.1039497 1.22311958 0.3019160
Age 0.2470919 1 0.2470919 0.09736724 0.7552820
sex 3.2949805 1 3.2949805 1.29839578 0.2556377
bmi 3.0321278 1 3.0321278 1.19481799 0.2754513
E4_Positive 2.2527945 1 2.2527945 0.88771963 0.3470395
Residuals 611.5934050 241 2.5377320
──────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
────────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 0.004574037 0.2567257 241.0000 0.01781683 1.0000000
- MCI -0.450628964 0.3005403 241.0000 -1.49939601 0.8104784
- DAT -0.401644007 0.3692471 241.0000 -1.08773768 1.0000000
SCD - MCI -0.455203001 0.2853988 241.0000 -1.59497153 0.6721748
- DAT -0.406218044 0.3520076 241.0000 -1.15400353 1.0000000
MCI - DAT 0.048984957 0.3772258 241.0000 0.12985579 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(ICAM1_ng_mL, na.rm = TRUE), Std=sd(ICAM1_ng_mL, na.rm = TRUE),
Max=max(ICAM1_ng_mL, na.rm = TRUE), Min=min(ICAM1_ng_mL, na.rm = TRUE))4.14. VCAM1_Diag
(VCAM1_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = VCAM1_ng_mL,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "VCAM1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,40, by = 10), limits = c(0, 40)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = VCAM1_ng_mL ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = VCAM1_ng_mL ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - VCAM1_ng_mL
───────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────
Diag 6.25139682 3 2.08379894 0.226559302 0.8778681
Age 12.68414522 1 12.68414522 1.379073108 0.2414182
sex 29.84363657 1 29.84363657 3.244724489 0.0729037
bmi 0.02490871 1 0.02490871 0.002708179 0.9585398
E4_Positive 20.95405901 1 20.95405901 2.278212585 0.1325129
Residuals 2216.61852533 241 9.19758724
───────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
────────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD 0.38981922 0.4887462 241.0000 0.79759028 1.0000000
- MCI 0.15831082 0.5721591 241.0000 0.27669020 1.0000000
- DAT 0.10962157 0.7029609 241.0000 0.15594262 1.0000000
SCD - MCI -0.23150839 0.5433332 241.0000 -0.42608915 1.0000000
- DAT -0.28019764 0.6701409 241.0000 -0.41811748 1.0000000
MCI - DAT -0.04868925 0.7181506 241.0000 -0.06779811 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(VCAM1_ng_mL, na.rm = TRUE), Std=sd(VCAM1_ng_mL, na.rm = TRUE),
Max=max(VCAM1_ng_mL, na.rm = TRUE), Min=min(VCAM1_ng_mL, na.rm = TRUE))NA4.15. CRP_Diag
(CRP_Diag_plot <- ggplot(DC_thesis_m,
aes(x = Diag, y = CRP_pg_ml,
colour = Diag, fill = Diag)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "CRP (pg/mL)") +
scale_y_continuous(breaks = seq(0,50000, by = 12500), limits = c(0, 50000)) +
scale_colour_manual(values = c("black","black", "black", "black")) +
scale_fill_manual(values = c("honeydew3","indianred1", "#1ca9c9", "springgreen4")) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = CRP_pg_ml ~ Diag +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = CRP_pg_ml ~ Diag,
postHocCorr = "bonf")
ANCOVA
ANCOVA - CRP_pg_ml
───────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────
Diag 2.862652e+8 3 9.542173e+7 2.904676703 0.0354632
Age 341406.12 1 341406.12 0.010392542 0.9188860
sex 85584.19 1 85584.19 0.002605218 0.9593349
bmi 1.858878e+8 1 1.858878e+8 5.658500660 0.0181509
E4_Positive 2.257763e+7 1 2.257763e+7 0.687272247 0.4079141
Residuals 7.917107e+9 241 3.285107e+7
───────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - Diag
───────────────────────────────────────────────────────────────────────────────────────────────
Diag Diag Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────
CN - SCD -714.7575 923.6792 241.0000 -0.7738157 1.0000000
- MCI -2891.3525 1081.3210 241.0000 -2.6739077 0.0480607
- DAT 152.6176 1328.5227 241.0000 0.1148777 1.0000000
SCD - MCI -2176.5951 1026.8430 241.0000 -2.1196960 0.2103258
- DAT 867.3751 1266.4964 241.0000 0.6848619 1.0000000
MCI - DAT 3043.9702 1357.2295 241.0000 2.2427822 0.1549231
───────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(Diag)%>%
summarise(Median=median(CRP_pg_ml, na.rm = TRUE), Std=sd(CRP_pg_ml, na.rm = TRUE),
Max=max(CRP_pg_ml, na.rm = TRUE), Min=min(CRP_pg_ml, na.rm = TRUE))5. Diagnosis + Tau scheme
5.1. YKL40_Diag_T
(YKL40_diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = YKL40_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "YKL40 (ng/mL)") +
scale_y_continuous(breaks = seq(0,1200, by = 200), limits = c(0,1200)) +
scale_x_discrete(labels = c('CN T-', 'CN T+', "SCD T-", "SCD T+", "MCI T-", "MCI T+",
'DAT T-','DAT T+')) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = YKL40_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = YKL40_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - YKL40_ng_ml
──────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 470320.898 7 67188.700 5.3155557 0.0000116
Age 382187.881 1 382187.881 30.2363487 < .0000001
sex 7830.977 1 7830.977 0.6195386 0.4320052
bmi 22115.698 1 22115.698 1.7496577 0.1871947
E4_Positive 13943.274 1 13943.274 1.1031058 0.2946553
Residuals 2995683.393 237 12640.014
──────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -88.376903 32.25941 237.0000 -2.73956937 0.1853558
- SCD_T- 17.844635 19.93727 237.0000 0.89503886 1.0000000
- SCD_T+ -87.314258 32.29694 237.0000 -2.70348364 0.2060435
- MCI_T- -11.974748 26.54286 237.0000 -0.45114763 1.0000000
- MCI_T+ -93.582554 27.94243 237.0000 -3.34912028 0.0264041
- DAT_T- 83.237001 45.75978 237.0000 1.81899928 1.0000000
- DAT_T+ -61.142003 29.56598 237.0000 -2.06798512 1.0000000
CN_T+ - SCD_T- 106.221538 32.65173 237.0000 3.25316760 0.0366107
- SCD_T+ 1.062645 40.79321 237.0000 0.02604957 1.0000000
- MCI_T- 76.402155 36.98550 237.0000 2.06573241 1.0000000
- MCI_T+ -5.205651 37.74582 237.0000 -0.13791331 1.0000000
- DAT_T- 171.613904 52.42830 237.0000 3.27330648 0.0342046
- DAT_T+ 27.234900 38.69489 237.0000 0.70383721 1.0000000
SCD_T- - SCD_T+ -105.158892 31.93702 237.0000 -3.29269602 0.0320277
- MCI_T- -29.819383 25.82160 237.0000 -1.15482299 1.0000000
- MCI_T+ -111.427188 27.17871 237.0000 -4.09979744 0.0015916
- DAT_T- 65.392366 45.07083 237.0000 1.45088012 1.0000000
- DAT_T+ -78.986638 28.66581 237.0000 -2.75542995 0.1768727
SCD_T+ - MCI_T- 75.339509 35.99921 237.0000 2.09281018 1.0000000
- MCI_T+ -6.268296 36.53935 237.0000 -0.17154920 1.0000000
- DAT_T- 170.551258 51.46383 237.0000 3.31400252 0.0297846
- DAT_T+ 26.172255 37.20325 237.0000 0.70349378 1.0000000
MCI_T- - MCI_T+ -81.607805 31.95467 237.0000 -2.55386126 0.3158608
- DAT_T- 95.211749 47.79055 237.0000 1.99227143 1.0000000
- DAT_T+ -49.167255 33.86228 237.0000 -1.45197698 1.0000000
MCI_T+ - DAT_T- 176.819554 48.58696 237.0000 3.63923864 0.0093980
- DAT_T+ 32.440551 33.09708 237.0000 0.98016346 1.0000000
DAT_T- - DAT_T+ -144.379003 49.35647 237.0000 -2.92522943 0.1057317
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(YKL40_ng_ml, na.rm = TRUE), Std=sd(YKL40_ng_ml, na.rm = TRUE),
Max=max(YKL40_ng_ml, na.rm = TRUE), Min=min(YKL40_ng_ml, na.rm = TRUE))NA5.2. AXL_Diag_T
(AXL_diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = AXL_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "AXL (ng/mL)") +
scale_y_continuous(breaks = seq(0,50, by = 10), limits = c(0,50)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = AXL_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = AXL_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - AXL_ng_ml
──────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 1206.372689 7 172.338956 8.40755430 < .0000001
Age 10.517162 1 10.517162 0.51307966 0.4745139
sex 1.675020 1 1.675020 0.08171581 0.7752352
bmi 45.601577 1 45.601577 2.22467251 0.1371513
E4_Positive 20.478533 1 20.478533 0.99904501 0.3185611
Residuals 4858.051587 237 20.498108
──────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -6.1576191 1.2990910 237.0000 -4.7399443 0.0001033
- SCD_T- 1.4713644 0.8028767 237.0000 1.8326155 1.0000000
- SCD_T+ -3.7831358 1.3006023 237.0000 -2.9087567 0.1112588
- MCI_T- 2.6805786 1.0688846 237.0000 2.5078279 0.3588855
- MCI_T+ -1.7550316 1.1252454 237.0000 -1.5596879 1.0000000
- DAT_T- 2.9361941 1.8427525 237.0000 1.5933741 1.0000000
- DAT_T+ -0.9806632 1.1906260 237.0000 -0.8236534 1.0000000
CN_T+ - SCD_T- 7.6289835 1.3148895 237.0000 5.8019960 0.0000006
- SCD_T+ 2.3744833 1.6427480 237.0000 1.4454337 1.0000000
- MCI_T- 8.8381977 1.4894113 237.0000 5.9340208 0.0000003
- MCI_T+ 4.4025875 1.5200292 237.0000 2.8963835 0.1155828
- DAT_T- 9.0938132 2.1112949 237.0000 4.3072207 0.0006783
- DAT_T+ 5.1769560 1.5582483 237.0000 3.3222920 0.0289522
SCD_T- - SCD_T+ -5.2545002 1.2861081 237.0000 -4.0855821 0.0016854
- MCI_T- 1.2092142 1.0398396 237.0000 1.1628854 1.0000000
- MCI_T+ -3.2263960 1.0944902 237.0000 -2.9478528 0.0985497
- DAT_T- 1.4648297 1.8150083 237.0000 0.8070650 1.0000000
- DAT_T+ -2.4520276 1.1543762 237.0000 -2.1241148 0.9715017
SCD_T+ - MCI_T- 6.4637144 1.4496930 237.0000 4.4586779 0.0003566
- MCI_T+ 2.0281042 1.4714445 237.0000 1.3783083 1.0000000
- DAT_T- 6.7193299 2.0724554 237.0000 3.2422072 0.0379853
- DAT_T+ 2.8024726 1.4981799 237.0000 1.8705848 1.0000000
MCI_T- - MCI_T+ -4.4356102 1.2868191 237.0000 -3.4469570 0.0187775
- DAT_T- 0.2556155 1.9245320 237.0000 0.1328196 1.0000000
- DAT_T+ -3.6612418 1.3636388 237.0000 -2.6849059 0.2174873
MCI_T+ - DAT_T- 4.6912257 1.9566037 237.0000 2.3976372 0.4837310
- DAT_T+ 0.7743684 1.3328240 237.0000 0.5809983 1.0000000
DAT_T- - DAT_T+ -3.9168572 1.9875919 237.0000 -1.9706547 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(AXL_ng_ml, na.rm = TRUE), Std=sd(AXL_ng_ml, na.rm = TRUE),
Max=max(AXL_ng_ml, na.rm = TRUE), Min=min(AXL_ng_ml, na.rm = TRUE))5.3. Tyro3_Diag_T
(Tyro3_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = Tyro3_pg_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Tyro3 (pg/mL)") +
scale_y_continuous(breaks = seq(0,10000, by = 1000), limits = c(0,10000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Tyro3_pg_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Tyro3_pg_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Tyro3_pg_ml
─────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 8.327400e+7 7 1.189629e+7 9.0301429 < .0000001
Age 4280910.4 1 4280910.4 3.2495213 0.0727147
sex 286891.7 1 286891.7 0.2177716 0.6411718
bmi 1206691.7 1 1206691.7 0.9159665 0.3395115
E4_Positive 3426971.6 1 3426971.6 2.6013198 0.1081055
Residuals 3.122232e+8 237 1317397.3
─────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -1737.42735 329.3374 237.0000 -5.2755237 0.0000084
- SCD_T- 249.86216 203.5403 237.0000 1.2275809 1.0000000
- SCD_T+ -686.10884 329.7205 237.0000 -2.0808798 1.0000000
- MCI_T- 598.47824 270.9769 237.0000 2.2085949 0.7884945
- MCI_T+ -608.05948 285.2652 237.0000 -2.1315588 0.9540258
- DAT_T- 1303.74065 467.1631 237.0000 2.7907614 0.1592264
- DAT_T+ 67.59165 301.8400 237.0000 0.2239320 1.0000000
CN_T+ - SCD_T- 1987.28951 333.3425 237.0000 5.9617039 0.0000003
- SCD_T+ 1051.31851 416.4592 237.0000 2.5244215 0.3428091
- MCI_T- 2335.90559 377.5862 237.0000 6.1864164 < .0000001
- MCI_T+ 1129.36787 385.3483 237.0000 2.9307718 0.1039292
- DAT_T- 3041.16800 535.2423 237.0000 5.6818533 0.0000011
- DAT_T+ 1805.01900 395.0374 237.0000 4.5692363 0.0002206
SCD_T- - SCD_T+ -935.97100 326.0461 237.0000 -2.8706711 0.1250627
- MCI_T- 348.61608 263.6136 237.0000 1.3224510 1.0000000
- MCI_T+ -857.92164 277.4683 237.0000 -3.0919628 0.0623421
- DAT_T- 1053.87849 460.1295 237.0000 2.2903952 0.6405711
- DAT_T+ -182.27051 292.6502 237.0000 -0.6228272 1.0000000
SCD_T+ - MCI_T- 1284.58708 367.5171 237.0000 3.4953125 0.0158214
- MCI_T+ 78.04936 373.0314 237.0000 0.2092300 1.0000000
- DAT_T- 1989.84948 525.3959 237.0000 3.7873334 0.0054059
- DAT_T+ 753.70048 379.8092 237.0000 1.9844188 1.0000000
MCI_T- - MCI_T+ -1206.53772 326.2263 237.0000 -3.6984684 0.0075488
- DAT_T- 705.26240 487.8953 237.0000 1.4455200 1.0000000
- DAT_T+ -530.88660 345.7012 237.0000 -1.5356807 1.0000000
MCI_T+ - DAT_T- 1911.80013 496.0259 237.0000 3.8542345 0.0041872
- DAT_T+ 675.65113 337.8892 237.0000 1.9996234 1.0000000
DAT_T- - DAT_T+ -1236.14900 503.8818 237.0000 -2.4532518 0.4165982
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(Tyro3_pg_ml, na.rm = TRUE), Std=sd(Tyro3_pg_ml, na.rm = TRUE),
Max=max(Tyro3_pg_ml, na.rm = TRUE), Min=min(Tyro3_pg_ml, na.rm = TRUE))NA5.4. TREM2_Diag_T
(TREM2_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = TREM2_pg_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TREM2 (pg/mL)") +
scale_y_continuous(breaks = seq(0,14000, by = 2000), limits = c(0,14000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TREM2_pg_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TREM2_pg_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TREM2_pg_ml
──────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 6.740926e+7 7 9629894.5 7.5703877 < .0000001
Age 1.855725e+7 1 1.855725e+7 14.5884855 0.0001708
sex 539198.0 1 539198.0 0.4238819 0.5156371
bmi 427507.6 1 427507.6 0.3360783 0.5626524
E4_Positive 3.396967e+7 1 3.396967e+7 26.7047160 0.0000005
Residuals 3.014753e+8 237 1272047.7
──────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -1271.64044 323.6193 237.0000 -3.9294334 0.0031291
- SCD_T- 416.50008 200.0063 237.0000 2.0824347 1.0000000
- SCD_T+ 112.23878 323.9958 237.0000 0.3464205 1.0000000
- MCI_T- 782.82584 266.2721 237.0000 2.9399471 0.1010070
- MCI_T+ -528.41708 280.3122 237.0000 -1.8851017 1.0000000
- DAT_T- 1181.02923 459.0519 237.0000 2.5727573 0.2995772
- DAT_T+ -447.94670 296.5993 237.0000 -1.5102755 1.0000000
CN_T+ - SCD_T- 1688.14052 327.5549 237.0000 5.1537640 0.0000151
- SCD_T+ 1383.87922 409.2284 237.0000 3.3816793 0.0235927
- MCI_T- 2054.46628 371.0304 237.0000 5.5371916 0.0000023
- MCI_T+ 743.22336 378.6577 237.0000 1.9627844 1.0000000
- DAT_T- 2452.66967 525.9491 237.0000 4.6633213 0.0001456
- DAT_T+ 823.69374 388.1785 237.0000 2.1219458 0.9766454
SCD_T- - SCD_T+ -304.26130 320.3851 237.0000 -0.9496737 1.0000000
- MCI_T- 366.32576 259.0366 237.0000 1.4141853 1.0000000
- MCI_T+ -944.91716 272.6507 237.0000 -3.4656687 0.0175770
- DAT_T- 764.52915 452.1405 237.0000 1.6909104 1.0000000
- DAT_T+ -864.44678 287.5691 237.0000 -3.0060492 0.0820789
SCD_T+ - MCI_T- 670.58706 361.1361 237.0000 1.8568820 1.0000000
- MCI_T+ -640.65586 366.5546 237.0000 -1.7477773 1.0000000
- DAT_T- 1068.79045 516.2737 237.0000 2.0702012 1.0000000
- DAT_T+ -560.18548 373.2147 237.0000 -1.5009736 1.0000000
MCI_T- - MCI_T+ -1311.24292 320.5622 237.0000 -4.0904476 0.0016527
- DAT_T- 398.20339 479.4242 237.0000 0.8305867 1.0000000
- DAT_T+ -1230.77254 339.6989 237.0000 -3.6231277 0.0099704
MCI_T+ - DAT_T- 1709.44631 487.4136 237.0000 3.5071778 0.0151658
- DAT_T+ 80.47038 332.0226 237.0000 0.2423642 1.0000000
DAT_T- - DAT_T+ -1628.97593 495.1332 237.0000 -3.2899753 0.0323252
──────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(TREM2_pg_ml, na.rm = TRUE), Std=sd(TREM2_pg_ml, na.rm = TRUE),
Max=max(TREM2_pg_ml, na.rm = TRUE), Min=min(TREM2_pg_ml, na.rm = TRUE))NA5.5. C1q_Diag_T
(C1q_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = C1q_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C1q (ng/mL)") +
scale_y_continuous(breaks = seq(0,600, by = 100), limits = c(0,600)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C1q_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C1q_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C1q_ng_ml
─────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 61479.1723 7 8782.7389 2.6755863 0.0110400
Age 27467.6097 1 27467.6097 8.3677725 0.0041748
sex 41731.8941 1 41731.8941 12.7132648 0.0004391
bmi 328.5387 1 328.5387 0.1000865 0.7520043
E4_Positive 404.7730 1 404.7730 0.1233106 0.7257832
Residuals 777963.7325 237 3282.5474
─────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -22.72467777 16.43948 237.0000 -1.38232301 1.0000000
- SCD_T- 10.60142472 10.16009 237.0000 1.04343823 1.0000000
- SCD_T+ -53.64005028 16.45861 237.0000 -3.25908762 0.0358875
- MCI_T- 2.10036197 13.52631 237.0000 0.15527971 1.0000000
- MCI_T+ -13.93767015 14.23954 237.0000 -0.97880080 1.0000000
- DAT_T- 2.07742950 23.31931 237.0000 0.08908625 1.0000000
- DAT_T+ -15.37347672 15.06690 237.0000 -1.02034754 1.0000000
CN_T+ - SCD_T- 33.32610249 16.63941 237.0000 2.00284186 1.0000000
- SCD_T+ -30.91537251 20.78833 237.0000 -1.48715050 1.0000000
- MCI_T- 24.82503974 18.84791 237.0000 1.31712408 1.0000000
- MCI_T+ 8.78700762 19.23537 237.0000 0.45681510 1.0000000
- DAT_T- 24.80210727 26.71760 237.0000 0.92830580 1.0000000
- DAT_T+ 7.35120105 19.71902 237.0000 0.37279752 1.0000000
SCD_T- - SCD_T+ -64.24147500 16.27519 237.0000 -3.94720261 0.0029191
- MCI_T- -8.50106275 13.15876 237.0000 -0.64603831 1.0000000
- MCI_T+ -24.53909486 13.85034 237.0000 -1.77173208 1.0000000
- DAT_T- -8.52399522 22.96821 237.0000 -0.37112137 1.0000000
- DAT_T+ -25.97490144 14.60818 237.0000 -1.77810718 1.0000000
SCD_T+ - MCI_T- 55.74041225 18.34529 237.0000 3.03840392 0.0740550
- MCI_T+ 39.70238013 18.62055 237.0000 2.13218083 0.9525779
- DAT_T- 55.71747978 26.22611 237.0000 2.12450448 0.9705801
- DAT_T+ 38.26657356 18.95888 237.0000 2.01839881 1.0000000
MCI_T- - MCI_T+ -16.03803212 16.28419 237.0000 -0.98488373 1.0000000
- DAT_T- -0.02293247 24.35419 237.0000 -9.416231e-4 1.0000000
- DAT_T+ -17.47383869 17.25631 237.0000 -1.01260571 1.0000000
MCI_T+ - DAT_T- 16.01509965 24.76005 237.0000 0.64681215 1.0000000
- DAT_T+ -1.43580657 16.86636 237.0000 -0.08512841 1.0000000
DAT_T- - DAT_T+ -17.45090622 25.15219 237.0000 -0.69381256 1.0000000
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(C1q_ng_ml, na.rm = TRUE), Std=sd(C1q_ng_ml, na.rm = TRUE),
Max=max(C1q_ng_ml, na.rm = TRUE), Min=min(C1q_ng_ml, na.rm = TRUE))5.6. C3_Diag_T
(C3_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = C3_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C3 (ng/mL)") +
scale_y_continuous(breaks = seq(0,5000, by = 500), limits = c(0, 5000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C3_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C3_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C3_ng_ml
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 1865439.6 7 266491.4 0.5512185 0.7948893
Age 498760.2 1 498760.2 1.0316500 0.3108072
sex 890439.0 1 890439.0 1.8418098 0.1760300
bmi 1234212.8 1 1234212.8 2.5528815 0.1114257
E4_Positive 100049.7 1 100049.7 0.2069457 0.6495884
Residuals 1.145797e+8 237 483458.7
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -164.479215 199.5090 237.0000 -0.82442021 1.0000000
- SCD_T- -92.432180 123.3024 237.0000 -0.74963786 1.0000000
- SCD_T+ 5.285473 199.7411 237.0000 0.02646163 1.0000000
- MCI_T- -204.080902 164.1548 237.0000 -1.24322216 1.0000000
- MCI_T+ -81.177746 172.8105 237.0000 -0.46975015 1.0000000
- DAT_T- -361.399822 283.0022 237.0000 -1.27702125 1.0000000
- DAT_T+ -218.813102 182.8513 237.0000 -1.19667207 1.0000000
CN_T+ - SCD_T- 72.047034 201.9352 237.0000 0.35678291 1.0000000
- SCD_T+ 169.764688 252.2864 237.0000 0.67290476 1.0000000
- MCI_T- -39.601687 228.7375 237.0000 -0.17313156 1.0000000
- MCI_T+ 83.301469 233.4397 237.0000 0.35684360 1.0000000
- DAT_T- -196.920608 324.2438 237.0000 -0.60732261 1.0000000
- DAT_T+ -54.333888 239.3092 237.0000 -0.22704466 1.0000000
SCD_T- - SCD_T+ 97.717653 197.5151 237.0000 0.49473513 1.0000000
- MCI_T- -111.648721 159.6942 237.0000 -0.69914074 1.0000000
- MCI_T+ 11.254434 168.0872 237.0000 0.06695592 1.0000000
- DAT_T- -268.967642 278.7414 237.0000 -0.96493620 1.0000000
- DAT_T+ -126.380922 177.2843 237.0000 -0.71287166 1.0000000
SCD_T+ - MCI_T- -209.366375 222.6378 237.0000 -0.94039014 1.0000000
- MCI_T+ -86.463219 225.9783 237.0000 -0.38261738 1.0000000
- DAT_T- -366.685295 318.2790 237.0000 -1.15208756 1.0000000
- DAT_T+ -224.098575 230.0842 237.0000 -0.97398508 1.0000000
MCI_T- - MCI_T+ 122.903156 197.6243 237.0000 0.62190310 1.0000000
- DAT_T- -157.318921 295.5616 237.0000 -0.53227125 1.0000000
- DAT_T+ -14.732201 209.4219 237.0000 -0.07034698 1.0000000
MCI_T+ - DAT_T- -280.222076 300.4870 237.0000 -0.93255977 1.0000000
- DAT_T+ -137.635356 204.6895 237.0000 -0.67241034 1.0000000
DAT_T- - DAT_T+ 142.586720 305.2460 237.0000 0.46712064 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(C3_ng_ml, na.rm = TRUE), Std=sd(C3_ng_ml, na.rm = TRUE),
Max=max(C3_ng_ml, na.rm = TRUE), Min=min(C3_ng_ml, na.rm = TRUE))5.7. C4_Diag_T
(C4_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = C4_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "C4 (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = C4_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = C4_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - C4_ng_ml
───────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 316562.98 7 45223.28 1.247260 0.2777083
Age 48632.08 1 48632.08 1.341275 0.2479749
sex 58051.09 1 58051.09 1.601052 0.2069968
bmi 94018.38 1 94018.38 2.593032 0.1086658
E4_Positive 39001.37 1 39001.37 1.075660 0.3007276
Residuals 8593167.35 237 36258.09
───────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -103.0116418 54.63679 237.0000 -1.88538980 1.0000000
- SCD_T- -25.7923252 33.76715 237.0000 -0.76382883 1.0000000
- SCD_T+ -67.0020054 54.70035 237.0000 -1.22489173 1.0000000
- MCI_T- -39.2395324 44.95483 237.0000 -0.87286572 1.0000000
- MCI_T+ -91.1896742 47.32524 237.0000 -1.92687188 1.0000000
- DAT_T- -100.7053384 77.50194 237.0000 -1.29939119 1.0000000
- DAT_T+ -102.1153983 50.07500 237.0000 -2.03924922 1.0000000
CN_T+ - SCD_T- 77.2193167 55.30123 237.0000 1.39633984 1.0000000
- SCD_T+ 36.0096365 69.09021 237.0000 0.52119737 1.0000000
- MCI_T- 63.7721095 62.64122 237.0000 1.01805340 1.0000000
- MCI_T+ 11.8219676 63.92894 237.0000 0.18492356 1.0000000
- DAT_T- 2.3063034 88.79622 237.0000 0.02597299 1.0000000
- DAT_T+ 0.8962435 65.53635 237.0000 0.01367552 1.0000000
SCD_T- - SCD_T+ -41.2096802 54.09076 237.0000 -0.76186180 1.0000000
- MCI_T- -13.4472072 43.73327 237.0000 -0.30748235 1.0000000
- MCI_T+ -65.3973490 46.03175 237.0000 -1.42070103 1.0000000
- DAT_T- -74.9130132 76.33509 237.0000 -0.98137066 1.0000000
- DAT_T+ -76.3230731 48.55041 237.0000 -1.57203756 1.0000000
SCD_T+ - MCI_T- 27.7624730 60.97076 237.0000 0.45534075 1.0000000
- MCI_T+ -24.1876688 61.88558 237.0000 -0.39084499 1.0000000
- DAT_T- -33.7033330 87.16272 237.0000 -0.38667143 1.0000000
- DAT_T+ -35.1133929 63.01001 237.0000 -0.55726689 1.0000000
MCI_T- - MCI_T+ -51.9501419 54.12066 237.0000 -0.95989485 1.0000000
- DAT_T- -61.4658060 80.94140 237.0000 -0.75938649 1.0000000
- DAT_T+ -62.8758660 57.35152 237.0000 -1.09632438 1.0000000
MCI_T+ - DAT_T- -9.5156642 82.29026 237.0000 -0.11563536 1.0000000
- DAT_T+ -10.9257241 56.05552 237.0000 -0.19490898 1.0000000
DAT_T- - DAT_T+ -1.4100599 83.59355 237.0000 -0.01686805 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(C4_ng_ml, na.rm = TRUE), Std=sd(C4_ng_ml, na.rm = TRUE),
Max=max(C4_ng_ml, na.rm = TRUE), Min=min(C4_ng_ml, na.rm = TRUE))5.8. FactorB_Diag_T
(FB_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = Factor_B_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Factor B (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Factor_B_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Factor_B_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Factor_B_ng_ml
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 253580.44 7 36225.78 0.8118664 0.5781757
Age 150781.90 1 150781.90 3.3792170 0.0672749
sex 14109.61 1 14109.61 0.3162145 0.5744230
bmi 302360.25 1 302360.25 6.7762831 0.0098205
E4_Positive 17781.16 1 17781.16 0.3984987 0.5284739
Residuals 1.057503e+7 237 44620.37
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -79.927352 60.61070 237.0000 -1.31870046 1.0000000
- SCD_T- -47.348458 37.45921 237.0000 -1.26400061 1.0000000
- SCD_T+ -82.430124 60.68121 237.0000 -1.35841272 1.0000000
- MCI_T- -87.943777 49.87013 237.0000 -1.76345594 1.0000000
- MCI_T+ -70.206046 52.49971 237.0000 -1.33726530 1.0000000
- DAT_T- -96.819433 85.97589 237.0000 -1.12612305 1.0000000
- DAT_T+ -85.629890 55.55013 237.0000 -1.54148869 1.0000000
CN_T+ - SCD_T- 32.578894 61.34779 237.0000 0.53105243 1.0000000
- SCD_T+ -2.502771 76.64444 237.0000 -0.03265431 1.0000000
- MCI_T- -8.016424 69.49032 237.0000 -0.11536030 1.0000000
- MCI_T+ 9.721307 70.91884 237.0000 0.13707651 1.0000000
- DAT_T- -16.892081 98.50507 237.0000 -0.17148438 1.0000000
- DAT_T+ -5.702537 72.70200 237.0000 -0.07843715 1.0000000
SCD_T- - SCD_T+ -35.081665 60.00496 237.0000 -0.58464609 1.0000000
- MCI_T- -40.595318 48.51500 237.0000 -0.83675811 1.0000000
- MCI_T+ -22.857587 51.06479 237.0000 -0.44761932 1.0000000
- DAT_T- -49.470975 84.68145 237.0000 -0.58420082 1.0000000
- DAT_T+ -38.281431 53.85885 237.0000 -0.71077333 1.0000000
SCD_T+ - MCI_T- -5.513653 67.63722 237.0000 -0.08151803 1.0000000
- MCI_T+ 12.224078 68.65206 237.0000 0.17805843 1.0000000
- DAT_T- -14.389310 96.69297 237.0000 -0.14881444 1.0000000
- DAT_T+ -3.199766 69.89944 237.0000 -0.04577671 1.0000000
MCI_T- - MCI_T+ 17.737731 60.03813 237.0000 0.29544108 1.0000000
- DAT_T- -8.875657 89.79142 237.0000 -0.09884749 1.0000000
- DAT_T+ 2.313887 63.62225 237.0000 0.03636915 1.0000000
MCI_T+ - DAT_T- -26.613388 91.28776 237.0000 -0.29153293 1.0000000
- DAT_T+ -15.423844 62.18455 237.0000 -0.24803338 1.0000000
DAT_T- - DAT_T+ 11.189544 92.73355 237.0000 0.12066338 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(Factor_B_ng_ml, na.rm = TRUE), Std=sd(Factor_B_ng_ml, na.rm = TRUE),
Max=max(Factor_B_ng_ml, na.rm = TRUE), Min=min(Factor_B_ng_ml, na.rm = TRUE))5.9. FactorH_Diag_T
(FH_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = Factor_H_ng_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "Factor H (ng/mL)") +
scale_y_continuous(breaks = seq(0,2000, by = 500), limits = c(0, 2000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = Factor_H_ng_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = Factor_H_ng_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - Factor_H_ng_ml
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 249103.633 7 35586.233 1.3041011 0.2490024
Age 11050.618 1 11050.618 0.4049634 0.5251511
sex 85340.020 1 85340.020 3.1273897 0.0782733
bmi 83663.299 1 83663.299 3.0659442 0.0812425
E4_Positive 8944.249 1 8944.249 0.3277730 0.5675161
Residuals 6467241.685 237 27287.940
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -64.874419 47.39888 237.0000 -1.36869102 1.0000000
- SCD_T- -13.551811 29.29391 237.0000 -0.46261529 1.0000000
- SCD_T+ -109.860784 47.45402 237.0000 -2.31509962 0.6009531
- MCI_T- -47.590290 38.99952 237.0000 -1.22027877 1.0000000
- MCI_T+ -60.429575 41.05591 237.0000 -1.47188475 1.0000000
- DAT_T- -31.158585 67.23501 237.0000 -0.46342796 1.0000000
- DAT_T+ -82.451040 43.44140 237.0000 -1.89798290 1.0000000
CN_T+ - SCD_T- 51.322608 47.97530 237.0000 1.06977140 1.0000000
- SCD_T+ -44.986364 59.93761 237.0000 -0.75055317 1.0000000
- MCI_T- 17.284129 54.34294 237.0000 0.31805657 1.0000000
- MCI_T+ 4.444844 55.46007 237.0000 0.08014494 1.0000000
- DAT_T- 33.715835 77.03310 237.0000 0.43767982 1.0000000
- DAT_T+ -17.576621 56.85454 237.0000 -0.30915069 1.0000000
SCD_T- - SCD_T+ -96.308972 46.92518 237.0000 -2.05239428 1.0000000
- MCI_T- -34.038479 37.93978 237.0000 -0.89717120 1.0000000
- MCI_T+ -46.877764 39.93377 237.0000 -1.17388763 1.0000000
- DAT_T- -17.606773 66.22273 237.0000 -0.26587204 1.0000000
- DAT_T+ -68.899229 42.11879 237.0000 -1.63583132 1.0000000
SCD_T+ - MCI_T- 62.270494 52.89377 237.0000 1.17727459 1.0000000
- MCI_T+ 49.431209 53.68740 237.0000 0.92072270 1.0000000
- DAT_T- 78.702199 75.61600 237.0000 1.04081414 1.0000000
- DAT_T+ 27.409744 54.66287 237.0000 0.50143254 1.0000000
MCI_T- - MCI_T+ -12.839285 46.95112 237.0000 -0.27346066 1.0000000
- DAT_T- 16.431705 70.21884 237.0000 0.23400709 1.0000000
- DAT_T+ -34.860750 49.75398 237.0000 -0.70066255 1.0000000
MCI_T+ - DAT_T- 29.270991 71.38901 237.0000 0.41002098 1.0000000
- DAT_T+ -22.021465 48.62967 237.0000 -0.45284013 1.0000000
DAT_T- - DAT_T+ -51.292455 72.51965 237.0000 -0.70729045 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(Factor_H_ng_ml, na.rm = TRUE), Std=sd(Factor_H_ng_ml, na.rm = TRUE),
Max=max(Factor_H_ng_ml, na.rm = TRUE), Min=min(Factor_H_ng_ml, na.rm = TRUE))5.10. MIF_Diag_T
(MIF_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = MIF_pg_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "MIF (pg/mL)") +
scale_y_continuous(breaks = seq(0,30000, by = 5000), limits = c(0, 30000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = MIF_pg_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = MIF_pg_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - MIF_pg_ml
───────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
───────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 5.926025e+8 7 8.465750e+7 8.31727399 < .0000001
Age 4.681935e+7 1 4.681935e+7 4.59982088 0.0329925
sex 967884.763 1 967884.763 0.09509096 0.7580730
bmi 708084.019 1 708084.019 0.06956653 0.7921981
E4_Positive 6323.295 1 6323.295 6.212394e-4 0.9801360
Residuals 2.412308e+9 237 1.017852e+7
───────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -3282.11706 915.4295 237.0000 -3.58533018 0.0114446
- SCD_T- 334.62461 565.7626 237.0000 0.59145769 1.0000000
- SCD_T+ -3967.68387 916.4945 237.0000 -4.32919560 0.0006185
- MCI_T- 350.92958 753.2101 237.0000 0.46591194 1.0000000
- MCI_T+ -2887.87124 792.9259 237.0000 -3.64204449 0.0093016
- DAT_T- 2938.85376 1298.5310 237.0000 2.26321413 0.6867816
- DAT_T+ -1830.24165 838.9975 237.0000 -2.18146248 0.8437090
CN_T+ - SCD_T- 3616.74167 926.5622 237.0000 3.90339868 0.0034629
- SCD_T+ -685.56681 1157.5940 237.0000 -0.59223423 1.0000000
- MCI_T- 3633.04664 1049.5424 237.0000 3.46155308 0.0178347
- MCI_T+ 394.24582 1071.1179 237.0000 0.36806950 1.0000000
- DAT_T- 6220.97082 1487.7646 237.0000 4.18142133 0.0011422
- DAT_T+ 1451.87541 1098.0497 237.0000 1.32223105 1.0000000
SCD_T- - SCD_T+ -4302.30848 906.2808 237.0000 -4.74721333 0.0001000
- MCI_T- 16.30497 732.7430 237.0000 0.02225196 1.0000000
- MCI_T+ -3222.49585 771.2536 237.0000 -4.17825712 0.0011571
- DAT_T- 2604.22914 1278.9806 237.0000 2.03617566 1.0000000
- DAT_T+ -2164.86626 813.4534 237.0000 -2.66132790 0.2328352
SCD_T+ - MCI_T- 4318.61345 1021.5541 237.0000 4.22749352 0.0009451
- MCI_T+ 1079.81263 1036.8817 237.0000 1.04140385 1.0000000
- DAT_T- 6906.53762 1460.3956 237.0000 4.72922368 0.0001084
- DAT_T+ 2137.44222 1055.7214 237.0000 2.02462723 1.0000000
MCI_T- - MCI_T+ -3238.80082 906.7819 237.0000 -3.57175292 0.0120224
- DAT_T- 2587.92417 1356.1586 237.0000 1.90827552 1.0000000
- DAT_T+ -2181.17123 960.9143 237.0000 -2.26989149 0.6751679
MCI_T+ - DAT_T- 5826.72500 1378.7585 237.0000 4.22606654 0.0009507
- DAT_T+ 1057.62959 939.2001 237.0000 1.12609609 1.0000000
DAT_T- - DAT_T+ -4769.09540 1400.5949 237.0000 -3.40504983 0.0217497
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(MIF_pg_ml, na.rm = TRUE), Std=sd(MIF_pg_ml, na.rm = TRUE),
Max=max(MIF_pg_ml, na.rm = TRUE), Min=min(MIF_pg_ml, na.rm = TRUE))NA5.11. TNFR1_Diag_T
(TNFR1_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = TNFR1_ng_mL,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TNFR1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,3, by = 0.5), limits = c(0, 3)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TNFR1_ng_mL ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TNFR1_ng_mL ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TNFR1_ng_mL
──────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 1.89190075 7 0.27027154 6.09731059 0.0000015
Age 0.13812558 1 0.13812558 3.11610539 0.0788096
sex 5.874071e-4 1 5.874071e-4 0.01325187 0.9084500
bmi 0.03331314 1 0.03331314 0.75154255 0.3868651
E4_Positive 4.959662e-8 1 4.959662e-8 1.118897e-6 0.9991569
Residuals 10.50534542 237 0.04432635
──────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -0.112821794 0.06041068 237.0000 -1.8675804 1.0000000
- SCD_T- 0.035230646 0.03733559 237.0000 0.9436211 1.0000000
- SCD_T+ -0.194149049 0.06048095 237.0000 -3.2100858 0.0422946
- MCI_T- 0.040976298 0.04970555 237.0000 0.8243807 1.0000000
- MCI_T+ -0.180594652 0.05232646 237.0000 -3.4513065 0.0184918
- DAT_T- 0.176306285 0.08569216 237.0000 2.0574377 1.0000000
- DAT_T+ -0.130506416 0.05536681 237.0000 -2.3571238 0.5384776
CN_T+ - SCD_T- 0.148052439 0.06114534 237.0000 2.4213201 0.4540559
- SCD_T+ -0.081327255 0.07639150 237.0000 -1.0646113 1.0000000
- MCI_T- 0.153798091 0.06926100 237.0000 2.2205583 0.7651620
- MCI_T+ -0.067772859 0.07068480 237.0000 -0.9588038 1.0000000
- DAT_T- 0.289128079 0.09818000 237.0000 2.9448777 0.0994680
- DAT_T+ -0.017684623 0.07246208 237.0000 -0.2440535 1.0000000
SCD_T- - SCD_T+ -0.229379695 0.05980694 237.0000 -3.8353358 0.0045022
- MCI_T- 0.005745652 0.04835489 237.0000 0.1188225 1.0000000
- MCI_T+ -0.215825298 0.05089627 237.0000 -4.2404931 0.0008956
- DAT_T- 0.141075639 0.08440200 237.0000 1.6714727 1.0000000
- DAT_T+ -0.165737062 0.05368111 237.0000 -3.0874375 0.0632613
SCD_T+ - MCI_T- 0.235125347 0.06741401 237.0000 3.4877816 0.0162511
- MCI_T+ 0.013554397 0.06842550 237.0000 0.1980898 1.0000000
- DAT_T- 0.370455334 0.09637387 237.0000 3.8439395 0.0043561
- DAT_T+ 0.063642633 0.06966876 237.0000 0.9135031 1.0000000
MCI_T- - MCI_T+ -0.221570950 0.05984000 237.0000 -3.7027229 0.0074302
- DAT_T- 0.135329988 0.08949510 237.0000 1.5121497 1.0000000
- DAT_T+ -0.171482714 0.06341229 237.0000 -2.7042504 0.2055830
MCI_T+ - DAT_T- 0.356900937 0.09098650 237.0000 3.9225702 0.0032140
- DAT_T+ 0.050088236 0.06197934 237.0000 0.8081441 1.0000000
DAT_T- - DAT_T+ -0.306812702 0.09242752 237.0000 -3.3194950 0.0292306
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(TNFR1_ng_mL, na.rm = TRUE), Std=sd(TNFR1_ng_mL, na.rm = TRUE),
Max=max(TNFR1_ng_mL, na.rm = TRUE), Min=min(TNFR1_ng_mL, na.rm = TRUE))5.12. TNFR2_Diag_T
(TNFR2_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = TNFR2_ng_mL,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "TNFR2 (ng/mL)") +
scale_y_continuous(breaks = seq(0,5, by = 1), limits = c(0, 5)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = TNFR2_ng_mL ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = TNFR2_ng_mL ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - TNFR2_ng_mL
─────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 5.38307453 7 0.76901065 4.8187094 0.0000430
Age 1.72842487 1 1.72842487 10.8305096 0.0011506
sex 0.09802654 1 0.09802654 0.6142456 0.4339763
bmi 0.04113729 1 0.04113729 0.2577710 0.6121278
E4_Positive 0.28739366 1 0.28739366 1.8008418 0.1808950
Residuals 37.82247638 237 0.15958851
─────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -0.19374397 0.11462610 237.0000 -1.69022563 1.0000000
- SCD_T- 0.02144822 0.07084232 237.0000 0.30276003 1.0000000
- SCD_T+ -0.33706850 0.11475945 237.0000 -2.93717430 0.1018821
- MCI_T- 0.06693815 0.09431369 237.0000 0.70973950 1.0000000
- MCI_T+ -0.34736954 0.09928672 237.0000 -3.49865053 0.0156343
- DAT_T- 0.21980007 0.16259640 237.0000 1.35181391 1.0000000
- DAT_T+ -0.26031693 0.10505562 237.0000 -2.47789635 0.3895890
CN_T+ - SCD_T- 0.21519219 0.11602008 237.0000 1.85478397 1.0000000
- SCD_T+ -0.14332453 0.14494888 237.0000 -0.98879364 1.0000000
- MCI_T- 0.26068212 0.13141912 237.0000 1.98359347 1.0000000
- MCI_T+ -0.15362557 0.13412071 237.0000 -1.14542766 1.0000000
- DAT_T- 0.41354404 0.18629141 237.0000 2.21987714 0.7664743
- DAT_T+ -0.06657296 0.13749300 237.0000 -0.48419165 1.0000000
SCD_T- - SCD_T+ -0.35851672 0.11348054 237.0000 -3.15927934 0.0500451
- MCI_T- 0.04548993 0.09175088 237.0000 0.49579824 1.0000000
- MCI_T+ -0.36881776 0.09657302 237.0000 -3.81905607 0.0047913
- DAT_T- 0.19835185 0.16014838 237.0000 1.23855047 1.0000000
- DAT_T+ -0.28176515 0.10185709 237.0000 -2.76627916 0.1712742
SCD_T+ - MCI_T- 0.40400665 0.12791456 237.0000 3.15841023 0.0501884
- MCI_T+ -0.01030104 0.12983381 237.0000 -0.07934022 1.0000000
- DAT_T- 0.55686857 0.18286438 237.0000 3.04525447 0.0724514
- DAT_T+ 0.07675157 0.13219283 237.0000 0.58060311 1.0000000
MCI_T- - MCI_T+ -0.41430769 0.11354328 237.0000 -3.64889674 0.0090699
- DAT_T- 0.15286192 0.16981227 237.0000 0.90018186 1.0000000
- DAT_T+ -0.32725508 0.12032150 237.0000 -2.71983866 0.1964218
MCI_T+ - DAT_T- 0.56716961 0.17264213 237.0000 3.28523298 0.0328499
- DAT_T+ 0.08705261 0.11760255 237.0000 0.74022725 1.0000000
DAT_T- - DAT_T+ -0.48011700 0.17537639 237.0000 -2.73763759 0.1864138
─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(TNFR2_ng_mL, na.rm = TRUE), Std=sd(TNFR2_ng_mL, na.rm = TRUE),
Max=max(TNFR2_ng_mL, na.rm = TRUE), Min=min(TNFR2_ng_mL, na.rm = TRUE))5.13. ICAM1_Diag_T
(ICAM1_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = ICAM1_ng_mL,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "ICAM1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,20, by = 5), limits = c(0, 20)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = ICAM1_ng_mL ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = ICAM1_ng_mL ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - ICAM1_ng_mL
─────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
─────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 40.02622617 7 5.71803231 2.33297054 0.0255041
Age 0.20714401 1 0.20714401 0.08451524 0.7715235
sex 4.52168938 1 4.52168938 1.84485983 0.1756739
bmi 7.11782822 1 7.11782822 2.90409054 0.0896657
E4_Positive 0.03046659 1 0.03046659 0.01243044 0.9113207
Residuals 580.87902777 237 2.45096636
─────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -0.22645109 0.4492120 237.0000 -0.5041074 1.0000000
- SCD_T- 0.11468180 0.2776263 237.0000 0.4130797 1.0000000
- SCD_T+ -0.69896626 0.4497346 237.0000 -1.5541749 1.0000000
- MCI_T- -0.05946476 0.3696091 237.0000 -0.1608856 1.0000000
- MCI_T+ -0.99180294 0.3890980 237.0000 -2.5489796 0.3201942
- DAT_T- 0.67383456 0.6372044 237.0000 1.0574857 1.0000000
- DAT_T+ -0.89949484 0.4117060 237.0000 -2.1847992 0.8367431
CN_T+ - SCD_T- 0.34113289 0.4546750 237.0000 0.7502786 1.0000000
- SCD_T+ -0.47251517 0.5680450 237.0000 -0.8318270 1.0000000
- MCI_T- 0.16698633 0.5150228 237.0000 0.3242310 1.0000000
- MCI_T+ -0.76535185 0.5256101 237.0000 -1.4561208 1.0000000
- DAT_T- 0.90028565 0.7300636 237.0000 1.2331606 1.0000000
- DAT_T+ -0.67304375 0.5388259 237.0000 -1.2490932 1.0000000
SCD_T- - SCD_T+ -0.81364806 0.4447227 237.0000 -1.8295629 1.0000000
- MCI_T- -0.17414656 0.3595656 237.0000 -0.4843249 1.0000000
- MCI_T+ -1.10648474 0.3784632 237.0000 -2.9236257 0.1062585
- DAT_T- 0.55915276 0.6276108 237.0000 0.8909228 1.0000000
- DAT_T+ -1.01417665 0.3991711 237.0000 -2.5407063 0.3276594
SCD_T+ - MCI_T- 0.63950150 0.5012886 237.0000 1.2757152 1.0000000
- MCI_T+ -0.29283668 0.5088101 237.0000 -0.5755324 1.0000000
- DAT_T- 1.37280082 0.7166333 237.0000 1.9156252 1.0000000
- DAT_T+ -0.20052859 0.5180549 237.0000 -0.3870798 1.0000000
MCI_T- - MCI_T+ -0.93233818 0.4449685 237.0000 -2.0952902 1.0000000
- DAT_T- 0.73329932 0.6654829 237.0000 1.1019055 1.0000000
- DAT_T+ -0.84003008 0.4715319 237.0000 -1.7814913 1.0000000
MCI_T+ - DAT_T- 1.66563750 0.6765730 237.0000 2.4618741 0.4069653
- DAT_T+ 0.09230810 0.4608765 237.0000 0.2002881 1.0000000
DAT_T- - DAT_T+ -1.57332940 0.6872884 237.0000 -2.2891838 0.6425713
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(ICAM1_ng_mL, na.rm = TRUE), Std=sd(ICAM1_ng_mL, na.rm = TRUE),
Max=max(ICAM1_ng_mL, na.rm = TRUE), Min=min(ICAM1_ng_mL, na.rm = TRUE))5.14. VCAM1_Diag_T
(VCAM1_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = VCAM1_ng_mL,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "VCAM1 (ng/mL)") +
scale_y_continuous(breaks = seq(0,40, by = 10), limits = c(0, 40)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = VCAM1_ng_mL ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = VCAM1_ng_mL ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - VCAM1_ng_mL
────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 136.667047 7 19.523864 2.2179797 0.0335701
Age 12.082916 1 12.082916 1.3726618 0.2425316
sex 35.499213 1 35.499213 4.0328357 0.0457577
bmi 4.532649 1 4.532649 0.5149249 0.4737213
E4_Positive 3.575452 1 3.575452 0.4061839 0.5245280
Residuals 2086.202875 237 8.802544
────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ -0.147772989 0.8513081 237.0000 -0.173583430 1.0000000
- SCD_T- 0.729271803 0.5261337 237.0000 1.386096090 1.0000000
- SCD_T+ -1.234155778 0.8522985 237.0000 -1.448032291 1.0000000
- MCI_T- 0.951317378 0.7004514 237.0000 1.358148970 1.0000000
- MCI_T+ -0.823264787 0.7373853 237.0000 -1.116464917 1.0000000
- DAT_T- 2.281165620 1.2075753 237.0000 1.889046323 1.0000000
- DAT_T+ -0.821064804 0.7802299 237.0000 -1.052337071 1.0000000
CN_T+ - SCD_T- 0.877044792 0.8616610 237.0000 1.017853595 1.0000000
- SCD_T+ -1.086382790 1.0765102 237.0000 -1.009170902 1.0000000
- MCI_T- 1.099090366 0.9760271 237.0000 1.126085958 1.0000000
- MCI_T+ -0.675491799 0.9960913 237.0000 -0.678142459 1.0000000
- DAT_T- 2.428938609 1.3835540 237.0000 1.755579190 1.0000000
- DAT_T+ -0.673291815 1.0211367 237.0000 -0.659355229 1.0000000
SCD_T- - SCD_T+ -1.963427582 0.8428003 237.0000 -2.329647481 0.5786415
- MCI_T- 0.222045574 0.6814179 237.0000 0.325858142 1.0000000
- MCI_T+ -1.552536591 0.7172311 237.0000 -2.164625445 0.8796281
- DAT_T- 1.551893817 1.1893942 237.0000 1.304776629 1.0000000
- DAT_T+ -1.550336607 0.7564750 237.0000 -2.049422199 1.0000000
SCD_T+ - MCI_T- 2.185473156 0.9499993 237.0000 2.300499872 0.6240985
- MCI_T+ 0.410890991 0.9642532 237.0000 0.426123530 1.0000000
- DAT_T- 3.515321398 1.3581021 237.0000 2.588407390 0.2866640
- DAT_T+ 0.413090974 0.9817732 237.0000 0.420760068 1.0000000
MCI_T- - MCI_T+ -1.774582165 0.8432662 237.0000 -2.104415057 1.0000000
- DAT_T- 1.329848242 1.2611663 237.0000 1.054459091 1.0000000
- DAT_T+ -1.772382182 0.8936070 237.0000 -1.983402386 1.0000000
MCI_T+ - DAT_T- 3.104430407 1.2821832 237.0000 2.421206653 0.4541941
- DAT_T+ 0.002199983 0.8734138 237.0000 0.002518833 1.0000000
DAT_T- - DAT_T+ -3.102230424 1.3024901 237.0000 -2.381768973 0.5045598
─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(VCAM1_ng_mL, na.rm = TRUE), Std=sd(VCAM1_ng_mL, na.rm = TRUE),
Max=max(VCAM1_ng_mL, na.rm = TRUE), Min=min(VCAM1_ng_mL, na.rm = TRUE))NA5.15. CRP_Diag_T
(CRP_Diag_T_plot <- ggplot(DC_thesis_m,
aes(x = clinical_N_cat, y = CRP_pg_ml,
colour = clinical_N_cat, fill = clinical_N_cat)) +
geom_boxplot() +
geom_jitter(pch = 18, size = 1, alpha = 0.8, width = 0.2) +
labs(x = NULL, y = "CRP (pg/mL)") +
scale_y_continuous(breaks = seq(0,50000, by = 12500), limits = c(0, 50000)) +
scale_colour_manual(values = c("black",'black', "black", "black",
"black", "black","black", 'black')) +
scale_fill_manual(values = c("honeydew3",'bisque2', "lightblue", "darkorange3",
"lightgreen", "coral","aquamarine2", 'deeppink')) +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank()))
ancova(formula = CRP_pg_ml ~ clinical_N_cat +
Age +
sex +
bmi +
E4_Positive,
data = DC_thesis_m,
postHoc = CRP_pg_ml ~ clinical_N_cat,
postHocCorr = "bonf")
ANCOVA
ANCOVA - CRP_pg_ml
──────────────────────────────────────────────────────────────────────────────────────
Sum of Squares df Mean Square F p
──────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat 4.120047e+8 7 5.885781e+7 1.790353210 0.0899112
Age 269993.5 1 269993.5 0.008212737 0.9278678
sex 2550218.6 1 2550218.6 0.077573258 0.7808563
bmi 1.592441e+8 1 1.592441e+8 4.843932380 0.0287085
E4_Positive 2.756193e+7 1 2.756193e+7 0.838386462 0.3607893
Residuals 7.791368e+9 237 3.287497e+7
──────────────────────────────────────────────────────────────────────────────────────
POST HOC TESTS
Post Hoc Comparisons - clinical_N_cat
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
clinical_N_cat clinical_N_cat Mean Difference SE df t p-bonferroni
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
CN_T- - CN_T+ 442.80197 1645.187 237.0000 0.26914986 1.0000000
- SCD_T- -1088.85127 1016.775 237.0000 -1.07088760 1.0000000
- SCD_T+ 1308.95361 1647.101 237.0000 0.79470132 1.0000000
- MCI_T- -1914.34671 1353.651 237.0000 -1.41421026 1.0000000
- MCI_T+ -3766.45201 1425.027 237.0000 -2.64307440 0.2453795
- DAT_T- 1210.32662 2333.688 237.0000 0.51863257 1.0000000
- DAT_T+ -96.76528 1507.826 237.0000 -0.06417538 1.0000000
CN_T+ - SCD_T- -1531.65324 1665.195 237.0000 -0.91980424 1.0000000
- SCD_T+ 866.15164 2080.400 237.0000 0.41633909 1.0000000
- MCI_T- -2357.14868 1886.212 237.0000 -1.24967341 1.0000000
- MCI_T+ -4209.25397 1924.987 237.0000 -2.18664055 0.8329205
- DAT_T- 767.52465 2673.774 237.0000 0.28705666 1.0000000
- DAT_T+ -539.56725 1973.388 237.0000 -0.27342179 1.0000000
SCD_T- - SCD_T+ 2397.80488 1628.746 237.0000 1.47217884 1.0000000
- MCI_T- -825.49544 1316.868 237.0000 -0.62686286 1.0000000
- MCI_T+ -2677.60074 1386.078 237.0000 -1.93178213 1.0000000
- DAT_T- 2299.17789 2298.552 237.0000 1.00027209 1.0000000
- DAT_T+ 992.08599 1461.918 237.0000 0.67861923 1.0000000
SCD_T+ - MCI_T- -3223.30032 1835.912 237.0000 -1.75569438 1.0000000
- MCI_T+ -5075.40562 1863.458 237.0000 -2.72364853 0.1942396
- DAT_T- -98.62699 2624.587 237.0000 -0.03757810 1.0000000
- DAT_T+ -1405.71890 1897.316 237.0000 -0.74089848 1.0000000
MCI_T- - MCI_T+ -1852.10530 1629.646 237.0000 -1.13650767 1.0000000
- DAT_T- 3124.67333 2437.255 237.0000 1.28204622 1.0000000
- DAT_T+ 1817.58142 1726.932 237.0000 1.05249185 1.0000000
MCI_T+ - DAT_T- 4976.77862 2477.871 237.0000 2.00848995 1.0000000
- DAT_T+ 3669.68672 1687.907 237.0000 2.17410430 0.8592474
DAT_T- - DAT_T+ -1307.09190 2517.115 237.0000 -0.51928182 1.0000000
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note. Comparisons are based on estimated marginal meansDC_thesis_m%>%
group_by(clinical_N_cat)%>%
summarise(Median=median(CRP_pg_ml, na.rm = TRUE), Std=sd(CRP_pg_ml, na.rm = TRUE),
Max=max(CRP_pg_ml, na.rm = TRUE), Min=min(CRP_pg_ml, na.rm = TRUE))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